Jump to ContentJump to Main Navigation
Language Policy and Political EconomyEnglish in a Global Context$

Thomas Ricento

Print publication date: 2015

Print ISBN-13: 9780199363391

Published to Oxford Scholarship Online: January 2015

DOI: 10.1093/acprof:oso/9780199363391.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 21 June 2018

The Economics of English in Europe

The Economics of English in Europe

Chapter:
(p.119) 5 The Economics of English in Europe
Source:
Language Policy and Political Economy
Author(s):

François Grin

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780199363391.003.0006

Abstract and Keywords

The economics of English constitute a special case, whether in general or in the European context. This chapter briefly reviews the field of “language economics” and discusses the implications of investigating the economics of a given language, as opposed to the economics of multilingualism or language in general. It then turns to the case of English, starting with the question of resource allocation, or “efficiency”. The data shows that learning English as a foreign language is a profitable investment, although other languages are equally useful and valuable on the labor market. English, therefore, cannot be construed as a “basic skill.” The chapter then addresses resource distribution, or “fairness”, showing that linguistic dominance (or hegemony) gives rise to considerable economic transfers, thus raising serious issues of linguistic justice. The concluding section discusses policy implications, in relation to the broader question of the respective values of linguistic uniformity and diversity.

Keywords:   economics of language, rates of return, efficiency, fairness, language policy, evaluation

Language Policy and Political Economy. Edited by Thomas Ricento.

© Oxford University Press 2015. Published 2015 by Oxford University Press.

1. Introduction

The economics of language studies the reciprocal influence between linguistic and economic processes. It has been an object of study for economists, albeit a relatively marginal one, since the 1960s. As the field has developed and branched out into the exploration of a broadening range of topics, a question that has emerged in recent years, against the backdrop of the set of processes commonly referred to as “globalization,” is whether these patterns of reciprocal influence present particular features in the case of English.1

The considerable and perhaps unprecedented weight of the English language internationally, at the turn of the twenty-first century, as well as the fact that this influence is occurring at a time of accelerated trade integration and development of information technology, leads us to consider the question from two different angles. First, we can assume that the interconnections between linguistic and economic processes are, in their essence, not vastly different for English than for other languages, and we would then mainly focus (data permitting) on the empirical estimation of the strength and magnitude of these interconnections. Second, starting out from the specific situation that the English language finds itself in, we can ask ourselves whether the study of the language–economics link requires, in this particular case, using or even developing specific concepts.

This chapter addresses both sets of issues and is organized as follows. Section 2 contains a general introduction to the economics of language (or language economics), providing an analytical perspective that serves, among other things, to distinguish between the economics of language in general, the economics of multilingualism (or linguistic diversity), and the economics of a particular language. Section 3 focuses on the economics of English in Europe from the perspective of resource allocation, whose pivotal concept (p.120) is that of “value.” The relative scarceness of suitable data, which hampers an extensive evaluation of the orders of magnitude associated with individual languages, including English, will lead us to rely not only on statistical results but on circumstantial evidence as well, in order to arrive at a more complete assessment of the value of English in contemporary Europe.

Owing to its focus on resource allocation, section 3 prioritizes the examination of what is known in economics as efficiency: The question there is whether scarce resources are used properly (as it were, in an economic way) and how much value their use generates. However, the proper use of scarce resources, though fundamental in economics, is not its only concern. Equally relevant to economic analysis is how the value added generated by (preferably efficient) resource allocation is distributed among members of society, which raises the issue of fairness. The economic perspective on fairness is less about ethical implications (which are studied in political philosophy) than about identifying the winners and losers from any given process (such as a public policy) and estimating the amounts gained and lost. This question is addressed in section 4 and as we shall see, to the extent that the case of English is a specific one from an economic standpoint, it is because of the distributive issues it raises, rather than for allocative reasons. In closing, section 5 considers the policy implications of the findings in sections 3 and 4.

Although this point will become clearer in due course, experience suggests that it deserves to be made at an early stage: Much of what will be said in the next pages does actually not concern English per se. It concerns a dominant or hegemonic language, and the policy implications derived in section 5 would be fundamentally the same, mutatis mutandis, if Spanish, Arabic, or Warlpiri found itself in this dominant position.

This chapter emphasizes the interpretation of facts in an economic perspective; as such, it does not offer a descriptive account of the frequency of use of English in Europe in different contexts. Basic demolinguistic figures regarding the distribution of English-language skills, however, can be found in Gazzola and Grin (2013), who use Eurobarometer data, or Gazzola (2014), who uses data from the much more robust Adult Education Survey (AES).

2. From Language Economics to the Economics of a Language

Let us start by briefly characterizing some essential features of language economics as a field of research. Readers interested in a more complete treatment may turn to analytical surveys in Grin (2003,2014) or Zhang and Grenier (2013).

(p.121) The early stages of language economics can be traced back to the 1960s, and since then, though still marginal in economics, it has evolved from a fringe interest with a focus on a relatively narrow selection of topics to a substantial research field addressing a wide range of issues. This evolution is largely due to the fact that language economics, as a field of research, has banked on the fundamentals of the discipline, emphasizing the fact that ultimately, economics is a tool for weighing the advantages and drawbacks of competing options, in order to identify the “optimal” one—namely, that which promises to deliver the best use of scarce resources. This problem is known as “optimal resource allocation.” When applied to language, therefore, economics emerges as a useful instrument for the selection, design, and evaluation of language policies. It provides a framework for identifying, processing, and organizing information about the positive effects (the advantages or benefits) as well the negative effects (drawbacks or costs) of possible policy scenarios.

Note that this exercise requires taking a broad view of what counts as advantages or benefits on the one hand, and drawbacks or costs on the other hand. A proper weighing of the relative merits of competing language policies must include not just the material and financial but also the nonmaterial or symbolic effects associated with each of them. A commonly made (an analytically very similar) distinction is between “market” and “nonmarket” effects. Economics is fundamentally concerned with both types of effects, since economic theory assigns no a priori restrictions on the nature of the resources used or the goals pursued. In other words, the nonmarket effects of a language policy are legitimate, even necessary components of the evaluation, alongside its market effects. Unfortunately, hard data on nonmarket effects are usually unavailable. The fallback option, then, is to qualify results obtained on market effects with circumstantial information suggesting orders of magnitude for the nonmarket ones.

Interdisciplinarity is an important feature of most language economics. Other specialties such as the economics of education, culture, or the environment have developed and moved into mainstream economics while only maintaining rather loose ties with the disciplines (such as the education sciences, sociology, and ecology) that traditionally focus on these issues. Not so with language economics, in which the linkages with the language disciplines have tended to get stronger with time. It is possible to formulate some economic propositions about language without engaging in essential sociolinguistic debates on the nature of language. However, the treatment of many topics in language economics, particularly when variables like language learning or language use are treated as dependent variables that might be (p.122) influenced by economic processes, requires not only a firm anchoring in economics but also a solid grasp of what language is about. Moreover, since these mutual influences operate in a context that is strongly embedded in social, political, and cultural realities, language economics is at its most relevant when engaging with other disciplines in the social sciences and humanities. This means not only linguistics and its many specialties but also the education sciences, sociology, political science, psychology, international relations, history, and law.

Economic perspectives on language can be split into two groups. A small branch of the specialty focuses on language itself (in the singular) and is arguably closer to structural linguistics (see, e.g., Marschak 1965; Rubinstein 2000). This line of work investigates language (either as a tool for communication or in terms of its internal structure) as the result of a complex process which is, in turn, driven by forces that may be deemed “economic.” The internal development of language and the structure of language at a certain point in time is, in this view, the result of economic forces. Work on these questions, however, constitutes a minority branch of language economics, which we shall leave aside in the rest of this chapter. In the main, language economics is not particularly interested in language corpus. It takes languages and their morphosyntactic features as a given, recognizing the existence of distinct languages (“Chinese,” “Italian,” “German,” “Navajo”).2 Most language economics addresses questions that have to do with the respective position of languages (in the plural) with respect to one another. In other words, and referring here to the well-known distinction between corpus and status, most of language economics is about status in a broad sense. We shall, however, return to some corpus questions in the closing section.

Early work in language economics was spurred on by socioeconomic problems in the outside world: In the 1960s, economists were invited to study socioeconomic inequality between anglophones and francophones in Québec in order to assess if this inequality was really correlated with language. The answer: yes, it was—francophones did earn less than anglophones, even if they had equivalent education, comparable professional experience, and were working in the same economic sectors. This disadvantage could be observed even for francophones with a good command of English (Vaillancourt 1996). Thus, economists could establish that there was language-based discrimination. This research provided some of the backing for language legislation that aimed at redressing the balance, with the result (attributable in part to the language legislation put in place in Québec) that some thirty years later, these earnings differentials had vanished (Vaillancourt, Lemay, and Vaillancourt 2007).

(p.123) Likewise, it was not just academic interest, but also political concern over continuing earnings differentials between some groups of immigrants in the United States (particularly Hispanics/Latinos) and the majority “white” population that gave rise to a large body of econometric research in this country (Bloom and Grenier 1996), mostly in the 1970s and 1980s, after which US economists’ interest in language, with a few exceptions, seems to have waned. While most of this econometric work used US data, it has inspired similar studies in Germany, Israel, and Australia (see Chiswick and Miller 2007, for an extensive review).

In such work, therefore, the emphasis was very much on economic issues. Language, usually in the form of a somewhat crude estimate of language skills, was then little more than a variable that was not considered intrinsically interesting but was taken into account simply because it could explain, better than other variables could, what happened with economic variables like earnings.

In parallel to econometric (or statistical) work focusing on empirical questions, other economists were developing theoretical models to propose explanations of earnings differentials (see, e.g., work by Lang 1986; Migué 1970; Raynauld and Marion 1972). Identifying the causes of earnings differentials through formal models could help to design policies intended to eliminate language-based inequality. This tradition is still strong, with a steady production flow of estimations of language-based earnings differentials. It remains relevant to track down and measure language-based inequality, which may often be interpreted as linguistic injustice.

However, another way of looking at these earnings differentials has emerged since then. Bi- or multilingual persons may be earning more, all other things being equal, for legitimate reasons, just like people endowed with advanced computer skills can expect a higher labor income. Therefore, instead of looking at whether immigrants tend to earn less because they have inadequate skills in the dominant language, economists started using the same techniques to assess whether some people earn more because they have learned foreign languages. Putting it differently, earnings differentials can also be a labor market reward for a sensible investment. At this time, returns to second-language skills are regularly estimated for Québec, sometimes Canada as a whole; and there exist one-off studies, with more or less extensive samples, for a small number of countries, including Switzerland (Grin 1999), Ukraine (Kastoukievich 2003), Luxembourg (Klein 2004), Israel (Lang and Siniver 2006), and India (Azam, Chin, and Prakash, 2011). Clearly, because of its importance in international communication, the case of the English language figures quite prominently in these studies. Research on the value of (p.124) second- or foreign-language skills remains, however, relatively rare, because it requires data which few countries collect (let alone collect on a regular basis, and with an adequate degree of detail). Even fewer countries are in a position to compare the returns to skills in different foreign languages.

Already in the 1980s, however, a few papers in language economics had appeared which, in a sense, turned the specialty on its head by reversing the causal direction examined. Work by Hočevar (1975, 1983) and Carr (1985), for example, proposed to use economic analysis not in order to understand economic phenomena like “wage inequality” but in order to understand language-related processes like language decline and language spread.3 This development opened up a whole new range of issues. From the 1990s onwards, new studies started to appear, with language playing a much more central analytical role than in earlier work. This could take two forms: In some contributions, language is indeed the dependent (or “explained” variable) responding to changes in the level of economic variables. This approach has been used, for example, in theoretical models of minority-language dynamics on language use by bilinguals, whose linguistic behavior at time t affects the vitality of a language at time t+1 (Grin 1992). Second, some studies focus on language policy and the alternative policy scenarios are approached through the prism of economic rationality (Grin and Vaillancourt 1999; Pool 1991). Even in studies (which experienced a surge in popularity around the turn of the 1990s) on the role of language in local or regional economic development, where the dependent variable (“development”) is an economic one, the true focus of the investigation is on language and the ways in which language (whether language skills, language use, language attitudes, all possibly mediated through large-scale social, political, and economic processes) would affect society at large (see, e.g., Ó Cinnéide and Keane 1988).

In the wake of this evolution in language economics giving language a larger analytical role, more attention is now being paid to sociolinguistic realities, including in work that does focus on economic variables. Some mainstream economics models are being revisited and “augmented” through the explicit inclusion of linguistic variables—for example, in production theory, where the standard textbook model can be expanded in order to include the language skills of suppliers, the language profile of clients, and language use among workers (e.g., Grin, Sfreddo, and Vaillancourt 2010). Understanding the role of language in core economic processes opens new avenues for language policy. Language planning bodies can then learn to harness market forces for the purposes of language policy, instead of waging an exhausting battle against such forces, as often seems to have been the case.

(p.125) Let us also mention a neighboring body of work, anchored in management studies more than economics, on the implications of multilingualism for internal and external communication in multilingual companies (Harzing and Feely 2007; Luo and Shenkar 2006; Marschan-Piekkari, Welch, and Welch 1999).

3. English as a Special Case: Assessing the Evidence

As we have seen, “value” can take many different forms. “Nonmarket” value (a concept which converges, not perfectly but to a significant extent, with notions such as “symbolic” or “intrinsic” value) should be taken into account alongside “market” value.

In the case of nonmarket value (or, rather, “values,” considering how varied its manifestations can be), many conceptual issues are still unresolved, and data are almost completely absent—even in the case of English, which is probably the language most often examined in the language economics literature. For example, what is the value of the direct enjoyment that a native speaker of French or Japanese may derive, after having acquired a high level of competence in English, from accessing the work of a US novelist whose books have not, or not yet, been translated into French or Japanese, respectively? What is the benefit of direct interaction with speakers of English when travelling to Australia, Britain, or Canada (assuming these interlocutors to be English-monolingual)? At this time, research has no clear answer to these questions.

Although we might expect market value to be better known, precise results are surprisingly few. Let us consider what is arguably the core of market value, namely, earnings differentials. Earnings differentials are made up of the extra labor income accruing to people with certain skills (or other traits), holding the main determinants of income (typically, education and work experience) constant. As noted earlier, what is often missing is the necessary RAD data, where “RAD” stands for representative, in adequate numbers, and sufficiently detailed. Few countries have collected such data, and fewer still do so on a regular basis. Moreover, assessing the market value of a language is only possible if we have a point of comparison, that is, similarly computed estimates of the rate of return to skills in other languages.

Let us therefore consider (Table 5.1) the earnings differentials accruing to men living in Switzerland’s three main language regions (respectively French-, German-, and Italian-speaking). The original survey (Grin 2001) includes 2,400 respondents of both genders; it is representative in terms of age, gender, and language region. In what follows, we focus on the results for the male subsample, which are typically more robust from a statistical standpoint. These men speak the locally dominant language (French, German, or (p.126) Italian, depending on their region of residence), and Table 5.1 reports their additional earnings as a result of their competence in French, German, or English as a foreign or second language, holding education, experience, and experience squared constant.4 This means, in essence, that the econometric procedure standardizes respondents in terms of education and experience, so that the earnings differences that still appear are not the result of education and experience. Therefore, the figures in Table 5.2 can be interpreted as the additional labor income (in percentage terms) accruing to people who have acquired an (approximately) B2 level in the foreign language concerned, as opposed to lower skills or no skills at all in that language.

Table 5.1 Language-Based Earnings Differentials (%)*

Switzerland, men, 1994/1995

L2→

Language region ↓

French

German

English

French-speaking region

13.8

10.2

German-speaking region

14.1

18.1

Italian-speaking region

17.2

16.9

ns

Let us make a few observations. First, language skills are eminently profitable. There are not many investments that result, for a given level of education and work experience, in wage premiums in the 10% to 18% range. Second, (p.127) these rates of profitability vary. In the two main language regions, they are remarkably symmetrical for Switzerland’s official languages, with a premium of about 14% for French in German-speaking Switzerland as well as for German in French-speaking Switzerland. The case of English is interesting, because it proves much more profitable in the German- than in the French-speaking region of the country, with premiums of 18% and 10%, respectively. That said, a 10% premium still amounts to a very attractive investment proposition. What is more, the rates of return for English are extremely robust in statistical terms, as shown by tests not reported here (detailed provided in Grin 1999).

The foregoing are examples of “private market value,” that is, value accruing to an individual person. But what is the value of investing in English from the standpoint of society as a whole? To answer this question, we need to move from “private value” to what is known in the literature as “social value.” Computing social value requires combining the econometric results on earnings differentials with data on spending on second-/foreign-language teaching and learning. Such data are difficult to come by, because education systems typically do not have numbers on spending by school subject, and estimates have to be derived from other school statistics. The resulting rates of return confirm that, in the Swiss case at least (i.e., given the amount that residents spend on language teaching and learning), foreign- or second-language skills constitute a highly profitable investment proposition for society as a whole, as shown in Table 5.2.

Table 5.2 Social Rates of Return on Foreign-Language Teaching (%)*

Switzerland, men, 1994/1995

L2 →

Language region ↓

French

German

English

French-speaking region

6.5

4.7

German-speaking region

10.0

12.6

Italian-speaking region

21.5

11.7

ns

*Control variables: education (years), experience (years), experience squared. Source: Grin (1999: 124). The private earnings differentials used as a basis for the calculations of social rates of return are significant at the 5% level for all reported coefficients.

There again, as for private returns, English does not necessarily guarantee the highest rates of social return; it remains profitable (the lowest figure in Table 5.2, at 4.7%, still represents an attractive return in real terms), but the results draw our attention to the fact that even if we focus on narrowly defined monetary benefits, English is not the only investment worth considering in a non-English-speaking country.

Extensions to the foregoing results, as well as estimates of other aspects of economic value, are available in the literature. For example, the macroeconomic counterpart of the aforementioned figures (specifically, the share of Switzerland’s GDP that can be traced back to these second-/foreign-language skills, including not just English but also a national language like German or French) is in the region of 9% to 10%; this high rate reflects, among others, the fact that in the Swiss economy, many languages are used, also for domestic trade across internal language boundaries. Similar calculations for Québec, where the majority language is French and where the use of another language means, most of the time, English, yield estimates in the 3–4% range.

The literature on various countries also contains results on:

  • (p.128) the impact of foreign language skills on access to employment (they generally improve it);

  • the likelihood of keeping one’s job when wages go up and firms try to reduce staff costs (bilinguals are about two to three times less likely to lose their job than monolinguals, other things being equal);

  • the actual usefulness of skills in given languages for particular occupations (the picture is a highly diverse one, with some sectors and jobs requiring high-level foreign-language skills, often in English, and others where other foreign languages matter more, and others still where they are not particularly relevant).

Overall, these findings tell us a fairly consistent story: that skills in “big” languages, through one or another channel, contribute to prosperity—but that this applies to many “big” languages, not to English only. These results can help to go beyond widespread, but somewhat clichéd, views. For example, the often-heard pronouncement that “English is the language of business,” which reflects the correct, if informal, observation that English is overproportionately used as the default language between trading partners who do not have the same mother tongue is, incorrectly, taken to imply that it is the only foreign language that leads to prosperity. This is simply not the case. Its usefulness and relevance are undisputed, but it is not the only linguistic asset worth investing in.5

This assessment is confirmed by converging evidence on facets of the role of language in economic life other than its impact on earnings. For example, 357 firms polled in the mid-1990s in the Brussels area report that what they most desperately need is not a workforce with skills in French and English or Dutch and English but with skills in French and Dutch, with English an added plus (Mettewie 2006). Similarly, a representative sample of firms in the manufacturing sector polled in German- and French-speaking Switzerland in 2005 indicate that they miss staff with skills in the national languages far more than staff with English-language skills (Grin, Sfreddo, and Vaillancourt 2010). In Britain itself, the Confederation for British Industry (CBI 2012) notes that over 70% of employers value foreign-language skills, and 21% express fear that business opportunities may be lost for lack of such skills. The English education system is known for its chronic underinvestment in this area, and the British Academy warns that “the lack of language skills at secondary, tertiary and research levels will affect the UK’s ability to compete effectively in a global market” (British Academy 2009).

(p.129) Of the dimensions that need to be taken into account for a more complete treatment of the question of value, time is of particular interest. More precisely, are we interested in the value of English now or its value in the long run? The figures presented earlier take into account the passage of time only in a limited, essentially technical way. What they do not (and cannot) tell us, however, is whether private earnings differentials and social rates of return estimated at time t will still hold n periods from now, at time t+n. On this matter, we only have circumstantial evidence, which suggests that as skills in a given language become more widespread, they also get more banal. Consequently, and even if the increase in the supply of skills is a response to the increase in the demand for those skills, they command an ever-decreasing premium in those sectors of the economy where they were highly profitable (other sectors in which they never were particularly useful remaining unaffected). In other words, although English is highly useful in many professions, it is likely that with the passage of time and the banalization of English, it is additional skills in other languages that will give people, and the economies in which they work, a true edge.

It is tempting to construe this last observation as implying that English should be regarded as a “basic” skill, a little like reading and writing: There was indeed a time when the ability to read and write was the preserve of the educated few, commanding, at least to some extent, access to socioeconomic prestige and privilege. Nowadays, reading and writing are indispensable for all, but these skills are not sufficient. Could the same be said of English? Probably not, because there is an essential difference between the two types of skills: Whereas it is almost impossible to identify one job or profession in which the ability to read and write is largely useless (implying that reading and writing are basic skills), a high number of professions, even in advanced, prosperous economies, require very little use, if any at all, of foreign languages. In the case of Switzerland (which was ranked by the World Bank, in 2012, the fifth country in the world in terms of per-capita GDP, at purchasing power parity), and even with the strong internal diversity of the country, “daily or almost daily” use of German (in the French-speaking region), French (in the German-speaking region), or English (in both) is reported by about one-third of respondents (percentages vary from 27.5% to 35.9%, where “use” can be oral or written, productive or receptive). For most people, the use of other languages is an infrequent occurrence, for which they can rely on the help of a colleague. By contrast, use of the written word, for reading or writing, certainly is a daily necessity for well over 90% of people.

The available evidence suggests, therefore, that foreign-language skills (i.e., skills in languages other than the local majority language) cannot be (p.130) construed as a “basic skill.” Majority-language skills can be considered a necessary basic skill in many cases: in English for Polish citizens moving to Britain for several years; in German for Turkish-speaking workers settling in Germany; and perhaps, in the future, in Putonghua for English-speaking immigrants moving to China. In fact, this may already be the case, according to the International Herald Tribune of May 23, 2013 (23), which reports on the difficulties encountered by high-level Western professionals seeking employment in the Far East: they may have fluent English, but those who do not have some competence in Chinese see the good jobs pass them by.

Summing up, the picture proves to be, under closer scrutiny, at variance with commonly held beliefs. English is extremely valuable, but it is not the only language that is valuable. Even in the narrowest, hardest-nosed approaches to value, prosperity is associated with linguistic diversity, encompassing “big” languages rather than one particular language, whether the latter is English (which is certainly the most influential language internationally at the beginning of the twenty-first century) or another language.

4. The Other Side of Economic Analysis: Distribution

Optimal resource allocation, which has been at the center of our attention in the preceding section, is only one side of economic analysis, even if it makes up, in practice, the clear majority of economic research. Equally relevant, however, is the problem of resource distribution, to which the founding fathers of the discipline assigned major importance (Bürgenmeier 1994). Any choice, including the choice of a language policy, impacts on the distribution of material and symbolic resources between social actors. We may investigate the distributive effects of the choices we make at different levels in order to assess who gains, who loses—and how much. And even if a policy only generates winners, there is still a difference between big winners and small winners, thus potentially reinforcing inequality in society. Distributive analysis focuses on identifying winners and losers, and estimating the amounts of the gains and losses they incur.

The identification of the actors affected can raise difficult problems. Sometimes it makes sense to consider distribution between individuals; at other times, it is more relevant to consider distribution between groups. Yet groups themselves can be defined according to very different criteria. Standard distributive analysis tends to reason in terms of socioeconomic status: For example, does a change in the tax system favor the poor, the middle-income groups, the well-off, or the super-rich? Depending on the question at hand, other groupings may be more relevant. A change in the legislation governing (p.131) retirement age and the financing of pension schemes usually needs to be assessed in terms of redistribution among age groups; child-care policies affect the distribution of resources between men and women, and so on. In other cases still, it is relevant to estimate how a particular ethnic group in society, often a minority, may be affected, by (usually implicit) comparison with the group considered as setting the norm; this often means the majority. In the case of language policies, the relevant groups are frequently defined in terms of linguistic attributes—that is, their first language (L1, also called “mother tongue”)6 and their skills in additional languages (L2, L3). While linguistic attributes belong to individuals, people with similar attributes constitute a group that is often distributionally relevant. Some forms of distributional analysis can also be carried out with respect to countries, to the extent that different countries are, generally, associated with different languages.

From a distributive perspective, therefore, the economics of the English language gives rise to the following question: How does the dominant position of English affect individuals, groups, or even countries depending on their linguistic attributes? For example, does the spread of English benefit the native speakers of English, or non-native but fluent speakers of it, or the nation-states conventionally lumped together under the label of “Anglo-Saxon countries”?

All these are immensely complex questions, for which we only have very partial answers. But before discussing the latter, let us note that from an analytical standpoint, the issues at hand cease to be about English proper (as opposed to other languages). The core issue is that of linguistic dominance, or even linguistic hegemony, irrespective of the language that finds itself in the role of hegemon. At the beginning of the twenty-first century, English is the language that happens to be in this role, but as already stressed earlier, the problem would be fundamentally the same if it were Éwé, Khmer, or Spanish.

A dominant language endows its speakers with a number of significant advantages.7 Although their estimation has only been the object of exploratory study, there is consensus on the notion that linguistic dominance implies at the very least a favorable position in communication (Ginsburgh and Weber 2011) that carries socioeconomic advantage, enough so that even staunch advocates of linguistic hegemony recognize that the dominance of English could justify compensatory payments to speakers of other languages (Van Parijs 2011).

Estimating the distributional effects of linguistic hegemony raises the complex problem of selecting a counterfactual, namely, an alternative state of affairs against which the observable one can be assessed. In the most general terms, (p.132) the logical counterfactual to hegemony is “diversity” or “multilingualism,” but the choice of a specific counterfactual will, in practice, depend on the issue at hand. Consider for example the formal institutions of the European Union (EU): the European Parliament uses all of the twenty-four official languages of the EU, in application of the EU’s oft-quoted “Regulation No. 1.”8 By contrast, the European Commission operates in three “procedural” languages; the Court of Justice has one official language (French), as does the European Central Bank (English). For a study of the distributional effects of the language regimes applying in the latter three bodies, a relevant counterfactual would be “use of all the official languages of Member states” (24). In other, less formalized contexts, the choice of a counterfactual is less self-evident, but a convenient solution is to use, as a counterfactual, a hypothetical state of affairs in which the presence and visibility of different languages is proportional to their demographic weight. Deviations from it are then interpreted as expressions of linguistic dominance, whose effects can then be estimated.9

A systematic treatment of the distributional issues at hand would be far beyond the scope of this chapter; moreover, theory does not yet provide an integrated analytical model for this purpose. Let us instead consider some essential facets of these distributional issues, as a succession of five vignettes on privileged markets, communication savings effort, language learning savings effort, knock-on effects, and legitimacy effects.

4.1 Privileged Markets

When a language is in a dominant or hegemonic position, its native speakers (and the national economy, or economies, in which these native speakers are concentrated) enjoy ipso facto advantages for the provision of certain goods and services. These include language courses, the associated teaching material and teacher training services, the provision of language learning environments (e.g., language stays of various forms), translation, and interpreting into the dominant language. No full-fledged estimates of the corresponding value for English are, to my knowledge, available.

Let us point out that it is not the entirety of the corresponding “language sector” that must be considered as distributionally problematic but only part of the “language sector.” The relevant part is what exceeds the volume of economic activity that would occur anyway, if English were one foreign language like another, with a share roughly proportional either to its demolinguistic weight or to this weight adjusted by a parameter reflecting the relative prosperity of the national economies associated with the English language.

(p.133) 4.2 Communication Savings Effort

Native speakers of the hegemonic language are spared the need to translate messages produced by speakers of other languages, since the latter will have made the effort to express themselves not in their own native language but in the hegemonic language. The same native speakers are also spared the effort to translate their own messages into other languages, expecting the others to have learned the hegemonic language. The corresponding economic effect can be approximated through the relative overrepresentation of untranslated and uninterpreted written and oral discourse in English in public- and private-sector contexts. To my knowledge, no solid basis for estimating these costs is available.

4.3 Language Learning Savings Effort

Native speakers of a hegemonic language hardly need to learn other languages, since their language is so frequently learned by others. Obviously, language learning is not costless, and the entirety of the cost of communicating between language spheres, in those communication contexts where translation and interpretation are not provided, is basically shunted onto the native speakers of languages other than the hegemonic one. Though obvious, the point needs to be made in passing: Reducing the number of official languages in the internal communication of some international organization does not save money. It simply replaces translation and interpreting costs within the organization with language learning costs outside the organization. The difference, however, is that whereas the former costs are shared by all members of the organization, the latter are borne only by the non-native speakers of the hegemonic language (Gazzola and Grin 2013).10

There is an easy rule of thumb to estimate the distributional effects that asymmetrical language learning effort entails. The (continental European) countries for which estimations of the expenditure for foreign-language teaching have been computed typically teach two foreign languages to over 95% of their school-going population. The corresponding expenditure averages 10% of total educational spending, of which roughly 6% is spent on the first foreign language, and 4% on the second foreign language taught (Grin 2005), simply because the first foreign language tends to have a greater weight in the syllabus. Whether the expenditure takes place in public or private schools makes little difference from the standpoint of aggregate financial flows. By having largely abandoned the requirement for schools to teach foreign languages at all, the English education system (as distinct from the Scottish and Welsh systems) saves roughly 6% of total spending. This, of course, is at (p.134) the expense of learners of English, in Europe and beyond, who learn it as a foreign or second language, at significant public, and often private, expense.

4.4 Knock-on Effects

All the amounts saved as part of the processes sketched out earlier can be invested in other pursuits, educational or otherwise. The return on the corresponding investment is part of the distributional effects of language hegemony. For example, money not invested in the teaching of foreign languages can be invested in basic numeracy or advanced computer skills, and at a very microlevel, scholars whose native language is the dominant one and who thus eschew any expenditure to get their papers published in international journals using the same language (advanced language learning; costly translation and text proofing) have more resources to devote to their core research topic. No estimates of such knock-on effects have, to my knowledge, been published to date.

4.5 Legitimization Effects

Let us finally turn to some effects that are arguably very important in actors’ daily life, but that are so multifaceted as to make systematic identification, let alone measurement, particularly challenging. In fact, these effects have never been evaluated. They have to do with the automatic advantage enjoyed by native speakers of a dominant language in most situations involving negotiation, competition, or conflict. As is well-known, mastery of the medium of communication means power (to set the agenda, to influence others), simply because the language itself carries authority (Bourdieu 1982). The same contents will sound more authoritative because they are expressed in the dominant language, and the opinions of those able to use it fluently are accorded more credence. What is more, the dominant language is construed as the norm, or default. Consequently, the use of any other language calls for explanation, justifications, and excuses (Phillipson 2010), while the use of the dominant language carries intrinsic legitimization—hence the label chosen here to describe this range of processes.

At the time of this writing, the study of legitimization effects is still fragmented. What makes them particularly complex is that they embody a tight interconnection between micro- and macro-level processes. However, illustrations abound, usually with reference to English. For example, we can observe that native speakers of English are overrepresented in a number of functions, not least in international organizations, including EU institutions, where in (p.135) 2010, eleven of twenty-six of the speakers of European Commissioners were native-born anglophones (representing a ratio of 42%, although native speakers of English make up less than 15% of the EU’s population).11 Some international corporations are imposing a certain degree of use of English on customers and employees, even in non-English-speaking countries.12 This gives rise to frequent situations of linguistic insecurity, in which non-native speakers, or even nonspeakers of English, find themselves forced to use the language—or to remain silent, when they are not simply excluded from communication. No method has yet, to my knowledge, been developed to estimate the level of these nonmarket costs and of their market consequences.

The legitimization effect is particularly useful for reminding us that a language cannot somehow be “neutral.” This has been amply discussed in the sociolinguistic literature (e.g. Wierzbicka 2014), and English can no more be neutral than any other unplanned language, unless of course native speakers of English did not exist. But they do, and the language, consequently, carries interests that are necessarily asymmetrical.

The examination of the economics of English yields a complex picture: From an allocative standpoint, it is a valuable asset to have—but so are other languages; from a distributive standpoint, English’s use in various settings, well over and above the demolinguistic weight of its native speakers, generates substantial material and symbolic inequalities. However, these would also occur no matter which language is in a dominant position.

5. Policy Implications

Our discussion so far has left out an important analytical distinction, which needs to be introduced now as we move on to a discussion of the policy implications that proceed (in Europe and beyond) from our economic examination of the role of English.

As we have seen, any economic assessment presupposes a point of reference, usually called the “counterfactual.” In our linguistically diverse world with its 6,000 to 7,000 languages, we can ask ourselves what foreign language investment is worthwhile, given that we live in a diverse world. This is what I call the “contingent” question, because it makes sense under a given contingence—precisely, the linguistic diversity that characterizes the world, and with which we are confronted with increasing frequency as a result of globalization. Our discussion so far has been framed in this “contingent” case, and as I have attempted to show, teaching and learning English (but not English only) are valuable.

On the other hand, we may question the value of linguistic diversity as such, the implicit counterfactual being linguistic uniformity. This is what (p.136) I call the “absolute” question, because it raises the issue of the value of diversity in the absolute. Mastering several languages is profitable, but all things considered, might not humankind be better off with one single language? Some commentators recommend precisely that, albeit in more muted terms, and they usually want this language to be English.13 They typically remain conveniently vague about the associated implications, and stop short of calling for the outright abandonment of other languages, hardly bothering to deny, however, that they would see nothing seriously wrong with the latter’s folklorization or with long-term linguistic uniformization—provided, of course, that uniformization takes place in English.14

Nevertheless, this debate raises a set of truly important questions: From the standpoint of economic analysis, should we prefer linguistic diversity or uniformity? If diversity is to be preferred, should the current spread of English, consequently, be countered? If, on the contrary, uniformity promises greater aggregate well-being, is English the right choice of language for the job? Let us address each of these questions in turn.15

The first raises matters of personal preference on which economic analysis, as such, has little to say. It can, however, be of help in offering an analytical framework for the systematic treatment of a question often approached in a surprisingly haphazard way. The main benefit of linguistic uniformity is generally unfettered communication. Its drawback is uniformity itself, which may detract from our quality of life (in the same way as environmental destruction can), and which entails the loss of the conceptual repertoires and cultural references associated with all the languages that would be sidelined. Moreover, converging circumstantial evidence suggests that linguistic diversity has positive impacts on creativity and innovation, which would be forfeited in a linguistically uniform world. Of course, considerable uncertainties surround the orders of magnitude concerned, both for benefits and costs, whether approached through uniformity or its obverse, diversity. But even advocates of uniformity can concede that when in doubt, we should err on the side of caution. It follows that a core objective for language policy is to nurture linguistic diversity rather than try to eliminate it (Diamond 2012).16

Turning now to the second question, if diversity is to be preserved, does the spread of a dominant language, ultimately drifting into hegemony, threaten it? The answer is most certainly yes, but the complex language dynamics involved cannot be further discussed here for lack of space. Let us simply signal the eviction effects that result from the spread of a dominant, and even more so a hegemonic, language.17 The mechanics of spread are probably nowhere more manifest than in academic teaching and research, where the use of English not in addition to but in displacement of other (p.137) languages carries serious allocative or distributive drawbacks (Gazzola 2012; Lévy-Leblond 1996; Phillipson [forthcoming]; Usunier 2010; ). This has particular relevance in Europe, which has been home to scientific research in a variety of languages from the earliest stages of science as a human endeavor. It follows that from an economic standpoint, it is wise to counter the processes abetting linguistic dominance and to support genuine multilingualism in science, teaching, and research.

One might think that the third question now becomes a moot one. If indeed diversity is preferable to uniformity, what is the point of discussing which language would be best suited to the task? The point is, ultimately, a geopolitical one: Advocates of linguistic uniformity are careful to disassociate themselves from any hint of being accessories to imperialism, insisting that they are only interested in the gains that humankind can reap from easier global communication (see Van Parijs 2012). If so, then, one would expect them to recommend whichever language is easiest to learn (for efficiency reasons) and minimizes unwarranted transfers between groups (for fairness reasons). From an economic standpoint, English is an ill-suited contender for the role of world language, because of its considerable morphosyntactic and phonological difficulties for the non-native speaker, and because it is the native language of some 400 million speakers, to whom massive advantages would then be offered, without compensation, but with colossal political and cultural consequences. It is very difficult to believe, therefore, that the voices calling for a greater role of English do so with no other goal in mind that the greater good of mankind.

This last observation will serve to close our discussion of the economics of English by stressing its limitations: Ultimately, the issues at hand are political more than economic. The economic perspective, however, can help by ensuring a better informed political debate.

Acknowledgement: The author thanks Marco Civico for helpful research assistance.

Notes

(1) In this chapter, I do not discuss the meaning of globalization or its positive or negative effects on linguistic diversity; though these effects should be considered ambiguous (Grin and Rossiaud, 1999), it is better to err on the side of caution and to remain alert to the risks of linguistic uniformity that globalization entails. At any rate, the issue still requires much more rigorous examination than has been the case so far (Ricento 2012).

(2) This is not to deny that languages are social constructions and that they lack sharp boundaries. However, from the rather pragmatic standpoint adopted by (p.138) economists, the fact remains that a person whose mother tongue is German and who has never studied Navajo will not understand a speech delivered in Navajo. For all the current vogue of the notion of “languaging,” there are distinct languages, such as German and Navajo (for a robust critique of “languaging,” see Edwards 2011: ch. 2).

(3) In a sense, these economists were returning to some of the early inspirations in language economics. These include, well before Marschak’s discussion of “economy” as a determinant of the features of language (1965), much earlier forerunners, with no less a figure than Adam Smith himself, one of the founding fathers of the discipline of economics. Smith had ventured some hypotheses about the connections between the development of language and the development of trade back in the 1740s, which have been taken up again and published in his Considerations Concerning the First Formation of Language, published in 1761.

(4) The presence of the term “experience squared” in the equation used in econometric estimation serves to take the obsolescence of skills into account; see Grin (1999) for detail.

(5) The dangers of clichés surrounding the English language in the language-economics connection may best be exemplified by an extra-European example. Let us consider the well-publicized results in the development economics literature, that insist that “linguistic fragmentation” (the negative pendant, as it were, of more positive “linguistic diversity”) has a negative impact on GDP per capita in developing economies. This result has received such publicity that it has become quasi-axiomatic in much development economics. However, an econometric investigation can be carried out more or less carefully, and the foregoing result is based on what is, upon closer examination, an insufficiently prudent application of basic methodology. It ignores the fact that both “English” (e.g., the presence of English-language skills in the resident population of the country considered) and GDP per capita can be influenced by another, typically unobserved factor (e.g., stable political institutions). When this possibility is taken into account using the relevant econometric procedures, “linguistic fragmentation” turns out not to have any statistically significant effect whatsoever; in those cases where it has one, this effect is actually positive. Interestingly, what applies to English applies to former colonial languages as well: There again, English is not the only language whose presence (or absence) generates certain economic effects (Arcand and Grin 2013).

(6) Some authors hold that languages, being constructs, do not really exist and “notions like ‘native speaker’ [and] ‘mother tongue’ . . . should have no place in the sociolinguistic toolkit itself” (Blommaert and Rampton 2011: 5). Such claims are squarely contradicted by overwhelming empirical evidence. Of course, it is true that for a (small) minority of language users, close association with more than one language and complex patterns of bilingualism may void the notion of one mother tongue; such patterns are typically associated with migration (p.139) during childhood, or with small, traditional societies with strongly exogamic conventions (Diamond 2012). This, however, is not true of the vast majority of the world’s population, whether made up of monolinguals or people for whom knowledge and use of one or more foreign language remains perfectly compatible with the concept and experience of “mother tongue.” This is confirmed by hard data where these are available. For example, in the last Canadian census (2011) where respondents could indicate as many mother tongues as they wished, 98% of them had no problem identifying one mother tongue. Similarly, 97% of respondents in a representative survey in Switzerland allowing five mother tongues to be mentioned gave just one (Grin 1999). On this point and on the continuing relevance of “named” languages, constructed as they may be, see May (2012) and Edwards (2011).

(7) In this section, I use the expressions “dominant” and “hegemonic” language equivalently, mainly to avoid repetition. Using these expressions helps to make it clear that the issue at hand is dominance or hegemony, not one particular language, whether it be English or another. In a more detailed discussion, however, the two terms need to be distinguished, with “hegemony” being a strong form of “dominance.”

(9) An analytically interesting counterfactual is to compare observable flows of written and oral messages (reflecting a situation in which English is in a dominant position) with those that would occur under a hypothetical regime centered on an external language, usually Esperanto or some other planned language. The advantage of using such an “external language counterfactual” is that it by referring to a situation in which the use of that sole language guarantees direct communication on a relatively equal footing (it being a language foreign to all), it allows the comparison to focus strictly on the effects of using an unplanned language (whether English, French, Chinese, or other) for the purposes of international communication.

(10) To the extent that language learning takes place in public schools financed out of general tax revenue, this cost is shouldered not by non-native speakers specifically, but by the entirety of the taxpayers in the society they belong to, including non-speakers of the language as well.

(12) Examples are particularly common in the telecommunications and travel industries, where instructions for the use of certain telephone or Internet services are often available in English only, although customers may not know the language. This raises issues of rights and safety, for both consumers and workers particularly in the health sector (Cuisiniez 2011).

(13) Others recommend Esperanto as the international language, but they never suggest that it should replace others in any function other than an auxiliary one.

(p.140) (14) Pronouncements of this kind abound in scholarly and media discourse since the early 1990s. Consider for example this gleeful excerpt from the Sunday Times of July 10, 1994: “Only when the French recognize the dominance of Anglo-American English as the universal language in a shrinking world can they effectively defend their own distinctive culture . . . . Britain must press ahead with the propagation of English and the British values which stand behind it” (quoted in Phillipson 2003: 4). In a more sober vein, a scholar like Coleman announces a future in which speakers will use “native languages for local and cultural communication where their personal identity is engaged, and another for international, formal, practical communication”—leaving no doubt in the reader’s mind that the language in question will be English (quoted in Phillipson [forthcoming]: 4).

(15) Before doing so, however, it is useful to put aside a notion that might otherwise obfuscate the discussion, namely, that of “English as a lingua franca,” or ELF. Advocates of ELF hold that English, when used by non-native speakers, serves purely instrumental purposes, and suddenly stops to be the vehicle of any cultural baggage, with the implication that non-native speakers are said to “own” ELF as much as native speakers of English own English. Quite apart from the wobbly character of the metaphor of “ownership,” no criterion is ever provided, in the abundant ELF literature, for distinguishing what counts as ELF (presented by some as a language different from English), and what does not, or what should or should not be considered an ELF interaction. This results in a hopelessly muddled vision whose main function seems to be what might be called “sanitization,” in the form of a crude syllogism going as follows: “Yes, the spread of English may be imperialistic; but ELF is not English; therefore the use of English in the form of ELF is not imperialistic.”

(16) Let us also point out that the shift to linguistic uniformity is not costless, which is why the compensatory payments mentioned before would need to be paid.

(17) Some authors, however, predict the natural retreat of dominant languages as a result of the development of information technologies and the decentralization that their use makes possible; see Ostler (2010).

References

Bibliography references:

Arcand, Jean-Louis and Grin, François (2013). “Language in Economic Development: Is English Special and Is Linguistic Fragmentation Bad?” in E. Erling and P. Seargeant (eds.), English and Development: Policy, Pedagogy and Globalization (Bristol, UK: Multilingual Matters), 243–266.

Azam, Mehtabul, Chin, Aimee and Prakash, Nishit, (2011). “The Returns to English-Language Skills in India”, Economic Development and Cultural Change 61/2: 335–367.

Blommaert, Jan and Rampton, Ben (2011). “Language and Superdiversity,” Diversities 13/2: 1–22.

(p.141) Bloom, David and Grenier, Gilles (1996). “Language, employment and earnings in the United States: Spanish-English differentials from 1970 to 1990”, International Journal of the Sociology of Language 121, 45–68.

Bourdieu, Pierre (1982). Langage et pouvoir symbolique (Paris: Arthème-Fayard).

British Academy (2009). Language Matters. A Position Paper. Available at http://www.britac.ac.uk/policy/language-matters/position-paper.cfm.

Bürgenmeier, Beat (1994). “The Misperception of Walras,” American Economic Review, 84 /1: 342–352.

Carr, Jack (1985). “Le bilinguisme au Canada: l’usage consacre-t-il l’anglais monopole naturel?” in F. Vaillancourt (ed.), Économie et langue (Québec: Conseil de la langue française), 27–37.

CBI (Confederation for British Industry) (2012). Learning to Grow. What Employers Need from Education and Skills. Education and Skills Survey 2012. Available at http://www.cbi.org.uk/media/1514978/cbi_education_and_skills_survey_2012.pdf.

Chiswick, Barry and Miller, Paul (2007). Economics of Language. International Analyses (New York: Routledge).

Cuisiniez, Jean-Loup (2011). “Le droit au français au travail face aux inconséquences des choix politiques. Analyse des actions syndicales dans le contexte français,” Français & Société, 22 /23: 33–48.

Diamond, Jared (2012). The World Until Yesterday (London: Allen Lane).

Edwards, John (2011). Multilingualism: Understanding Linguistic Diversity (London: Verso).

Gazzola, Michele (2012). “The Linguistic Implications of Academic Performance Indicators: General Trends and Case Study,” International Journal of the Sociology of Language, 216: 131–156.

——— (2014). “Partecipazione, esclusione linguistica e traduzione: Una valutazione del regime linguistico dell’Unione Europea.” Working paper No. 12, Observatoire Économie-Langues-Formation, University of Geneva, April 2014. Available at http://www.unige.ch/traduction-interpretation/recherches/groupes/elf/documents/elfwp12.pdf.

Gazzola, Michele and Grin, François (2013). “Is ELF More Effective and Fair Than Translation? An Evaluation of the EU’s Multilingual Regime,” International Journal of Applied Linguistics, 23 /1: 93–107.

Ginsburgh, Victor, and Weber, Shlomo (2011). How Many Languages Do We Need? The Economics of Linguistic Diversity (Princeton: Princeton University Press).

Grin, François (1992). “Towards a Threshold Theory of Minority Language Survival,” Kyklos, 45: 69–97.

——— (1999). Compétences et récompenses. La valeur des langues en Suisse (Fribourg: Éditions universitaires).

——— (2001). “English as Economic Value: Facts and Fallacies,” World Englishes, 20 /1: 65–78.

(p.142) ——— (2003). “Economics and Language Planning,” Current Issues in Language Planning, 4/1: 1–66.

——— (2005). L’enseignement des langues étrangères comme politique publique. Paris: La Documentation française. Available at http://www.ladocumentationfrancaise.fr/rapports-publics/054000678/.

——— (2014). “50 Years of Economics in Language Policy: Critical Assessment and Priorities,” ELFWorking Paper 13, University of Geneva, http://www.unige.ch/traduction-interpretation/recherches/groupes/elf/documents/elfwp13.pdf.

Grin, François and Jean Rossiaud (1999). “Mondialisation, processus marchands et dynamique des langues,” in S. Abou (ed.), Uniformisation ou différenciation des modèles culturels (Paris: AUPELF-UREF and Beirut: Université Saint-Joseph), 113–142.

Grin, François, Sfreddo, Claudio, and Vaillancourt, François (2010). The Economics of the Multilingual Workplace (New York: Routledge).

Grin, François and Vaillancourt, François (1999). The Cost-Effectiveness Evaluation of Minority Language Policies: Case Studies on Wales, Ireland and the Basque Country. Monograph No. 2. Flensburg: European Centre for Minority Issues. Available at http://edoc.vifapol.de/opus/volltexte/2008/796/pdf/monograph_2.pdf.

Harzing, Anne-Wil and Feely, Alan J. (2007). The Language Barrier and Its Implications for HQ-Subsidiary Relationships. Manuscript, Department of Management, Faculty of Economics and Commerce, University of Melbourne and Birmingham Business School.

Hočevar, Toussaint (1975). “Equilibria on linguistic minority markets,” Kyklos, 28: 337–357.

——— (1983): “Les aspects économiques de la dynamique fonctionnelle des langues,” Language Problems and Language Planning 7: 135–147.

Kastoukievitch, Nikolai (2003). Language Effects on Labor Market Outcomes in a Bilingual Economy: The Case of Ukraine. MA Dissertation, National University Kyiv-Mohyla, Kiev, Ukraine.

Klein, Carlo (2004). “La valorisation des compétences lingsuitiques: importance du sexe et/ou du statut professionnel?” Paper presented at the 11èmes Journées d’études sur les données longitudinales, Dijon (France), May 27–28.

Lang, Kevin (1986). “A Language Theory of Discrimination,” Quarterly Journal of Economics, 101: 363–382.

Lang, Kevin and Siniver, Erez (2006). “The Return to English in a non-English Speaking Country: Russian Immigrants and Native Israelis in Israel,” The B.E. Journal of Economic Analysis & Policy, 9/1: Art. 50.

Lévy-Leblond, Jean.-Marc (1996). La pierre de touche (la science à l’épreuve) (Paris: Seuil).

Luo, Yadong and Shenkar, Oded (2006). “The Multinational Corporation as a Multilingual Community: Language and Organization in a Global Context,” Journal of International Business Studies, 37/3: 321–340.

(p.143) Marschak, Jacob (1965). “Economics of Language,” Behavioral Science, 10: 135–140.

Marschan-Piekkari, Rebecca, Welch, Denice, and Welch, Lawrence (1999). “In the Shadow: The Impact of Language on Structure, Power and Communication in the Multinational,” International Business Review, 8: 421–440.

May, Stephen (2012). Language and Minority Rights. Ethnicity, Nationalism and the Politics of Language. 2nd ed. (New York: Routledge).

Mettewie, Laurence (2006). Entreprises bruxelloises et langues étrangères. Pratique et coût d’une main-d’œuvre ne maîtrisant pas les langues étrangères. Liège: TIBEM (Tweetaligheid in Beweging—Bilinguisme en Mouvement).

Migué, Jean-Luc (1970). “Le nationalisme, l’unité nationale et la théorie économique de l’information,” Revue canadienne d’économique, 3: 183–198.

Ó Cinnéide, Mícheál and Michael Keane (1988). Local Socio-economic Impacts Associated with the Galway Gaeltacht (Gaillimh/Galway, Eïre/Ireland: Coláiste na hOIllscoile Gaillimhe).

Ostler, Nicholas (2010). The Last Lingua Franca: English Until the Return of Babel (London: Allen Lane).

Phillipson, Robert (2003). English-Only Europe? (New York: Routledge).

——— (2010). Linguist Imperialism Continued (New York: Routledge).

——— (forthcoming). “English as Threat or Opportunity in European Higher Education,” in S. Dimova, A., K. Hultgren, and C. Jensen (eds.), English-Medium Instruction in Higher Education in Europe (Berlin: Mouton de Gruyter).

Pool, Jonathan (1991). “The Official Language Problem,” American Political Science Review, 85: 495–514.

Raynauld, André and Marion, Pierre (1972). “Une analyse économique de la disparité inter-ethnique des revenus,” Revue économique, 23: 1–19.

Ricento, Thomas (2012). “Political Economy and English as a ‘Global’ Language,” Critical Multilingualism Studies, 1: 30–52.

Rubinstein, Ariel (2000). Economics and Language (Cambridge: Cambridge University Press).

Usunier, Jean-Claude (2010). “Un plurilinguisme pragmatique face au mythe de l’anglais lingua franca de l’enseignement supérieur,” in Les enjeux du plurilinguisme pour la construction et la circulation des savoirs (Berne: SAGW/ASSH [Swiss Academy of the Social Sciences and Humanities]), 37–48.

Vaillancourt, François (1996). “Language and Socioeconomic Status in Quebec: Measurement, Findings, Determinants, and Policy Costs,” International Journal of the Sociology of Language, 121: 69–92.

Vaillancourt, François, Lemay, Dominique, and Vaillancourt, Luc (2007). “Laggards No More: The Changed Socioeconomic Status of Francophones in Quebec,” C.D. Howe Institute Backgrounder No. 103.

Van Parijs, Philippe (2011). Linguistic Justice for Europe and for the World (Oxford: Oxford University Press).

(p.144) Wassener, Bettina (2013, May) “Job Hunting in Asia: Language Skills Count,” International Herald Tribune. Available at http://rendezvous.blogs.nytimes.com/2013/05/23/job-hunting-in-asia-language-skills-count/?_php=true&_type=blogs&_r=0

Wierzbicka, Anna (2014). Imprisoned in English. The Hazards of English as a Default Language (Oxford: Oxford University Press).

Zhang, Wei Guo and Grenier, Gilles (2013). “How Can Language Be Linked to Economics? A Survey of Two Strands of Research” Language Problems and International Herald Tribune Language Planning, 37: 203–226.

Notes:

(1) In this chapter, I do not discuss the meaning of globalization or its positive or negative effects on linguistic diversity; though these effects should be considered ambiguous (Grin and Rossiaud, 1999), it is better to err on the side of caution and to remain alert to the risks of linguistic uniformity that globalization entails. At any rate, the issue still requires much more rigorous examination than has been the case so far (Ricento 2012).

(2) This is not to deny that languages are social constructions and that they lack sharp boundaries. However, from the rather pragmatic standpoint adopted by (p.138) economists, the fact remains that a person whose mother tongue is German and who has never studied Navajo will not understand a speech delivered in Navajo. For all the current vogue of the notion of “languaging,” there are distinct languages, such as German and Navajo (for a robust critique of “languaging,” see Edwards 2011: ch. 2).

(3) In a sense, these economists were returning to some of the early inspirations in language economics. These include, well before Marschak’s discussion of “economy” as a determinant of the features of language (1965), much earlier forerunners, with no less a figure than Adam Smith himself, one of the founding fathers of the discipline of economics. Smith had ventured some hypotheses about the connections between the development of language and the development of trade back in the 1740s, which have been taken up again and published in his Considerations Concerning the First Formation of Language, published in 1761.

(4) The presence of the term “experience squared” in the equation used in econometric estimation serves to take the obsolescence of skills into account; see Grin (1999) for detail.

(5) The dangers of clichés surrounding the English language in the language-economics connection may best be exemplified by an extra-European example. Let us consider the well-publicized results in the development economics literature, that insist that “linguistic fragmentation” (the negative pendant, as it were, of more positive “linguistic diversity”) has a negative impact on GDP per capita in developing economies. This result has received such publicity that it has become quasi-axiomatic in much development economics. However, an econometric investigation can be carried out more or less carefully, and the foregoing result is based on what is, upon closer examination, an insufficiently prudent application of basic methodology. It ignores the fact that both “English” (e.g., the presence of English-language skills in the resident population of the country considered) and GDP per capita can be influenced by another, typically unobserved factor (e.g., stable political institutions). When this possibility is taken into account using the relevant econometric procedures, “linguistic fragmentation” turns out not to have any statistically significant effect whatsoever; in those cases where it has one, this effect is actually positive. Interestingly, what applies to English applies to former colonial languages as well: There again, English is not the only language whose presence (or absence) generates certain economic effects (Arcand and Grin 2013).

(6) Some authors hold that languages, being constructs, do not really exist and “notions like ‘native speaker’ [and] ‘mother tongue’ . . . should have no place in the sociolinguistic toolkit itself” (Blommaert and Rampton 2011: 5). Such claims are squarely contradicted by overwhelming empirical evidence. Of course, it is true that for a (small) minority of language users, close association with more than one language and complex patterns of bilingualism may void the notion of one mother tongue; such patterns are typically associated with migration (p.139) during childhood, or with small, traditional societies with strongly exogamic conventions (Diamond 2012). This, however, is not true of the vast majority of the world’s population, whether made up of monolinguals or people for whom knowledge and use of one or more foreign language remains perfectly compatible with the concept and experience of “mother tongue.” This is confirmed by hard data where these are available. For example, in the last Canadian census (2011) where respondents could indicate as many mother tongues as they wished, 98% of them had no problem identifying one mother tongue. Similarly, 97% of respondents in a representative survey in Switzerland allowing five mother tongues to be mentioned gave just one (Grin 1999). On this point and on the continuing relevance of “named” languages, constructed as they may be, see May (2012) and Edwards (2011).

(7) In this section, I use the expressions “dominant” and “hegemonic” language equivalently, mainly to avoid repetition. Using these expressions helps to make it clear that the issue at hand is dominance or hegemony, not one particular language, whether it be English or another. In a more detailed discussion, however, the two terms need to be distinguished, with “hegemony” being a strong form of “dominance.”

(9) An analytically interesting counterfactual is to compare observable flows of written and oral messages (reflecting a situation in which English is in a dominant position) with those that would occur under a hypothetical regime centered on an external language, usually Esperanto or some other planned language. The advantage of using such an “external language counterfactual” is that it by referring to a situation in which the use of that sole language guarantees direct communication on a relatively equal footing (it being a language foreign to all), it allows the comparison to focus strictly on the effects of using an unplanned language (whether English, French, Chinese, or other) for the purposes of international communication.

(10) To the extent that language learning takes place in public schools financed out of general tax revenue, this cost is shouldered not by non-native speakers specifically, but by the entirety of the taxpayers in the society they belong to, including non-speakers of the language as well.

(12) Examples are particularly common in the telecommunications and travel industries, where instructions for the use of certain telephone or Internet services are often available in English only, although customers may not know the language. This raises issues of rights and safety, for both consumers and workers particularly in the health sector (Cuisiniez 2011).

(13) Others recommend Esperanto as the international language, but they never suggest that it should replace others in any function other than an auxiliary one.

(p.140) (14) Pronouncements of this kind abound in scholarly and media discourse since the early 1990s. Consider for example this gleeful excerpt from the Sunday Times of July 10, 1994: “Only when the French recognize the dominance of Anglo-American English as the universal language in a shrinking world can they effectively defend their own distinctive culture . . . . Britain must press ahead with the propagation of English and the British values which stand behind it” (quoted in Phillipson 2003: 4). In a more sober vein, a scholar like Coleman announces a future in which speakers will use “native languages for local and cultural communication where their personal identity is engaged, and another for international, formal, practical communication”—leaving no doubt in the reader’s mind that the language in question will be English (quoted in Phillipson [forthcoming]: 4).

(15) Before doing so, however, it is useful to put aside a notion that might otherwise obfuscate the discussion, namely, that of “English as a lingua franca,” or ELF. Advocates of ELF hold that English, when used by non-native speakers, serves purely instrumental purposes, and suddenly stops to be the vehicle of any cultural baggage, with the implication that non-native speakers are said to “own” ELF as much as native speakers of English own English. Quite apart from the wobbly character of the metaphor of “ownership,” no criterion is ever provided, in the abundant ELF literature, for distinguishing what counts as ELF (presented by some as a language different from English), and what does not, or what should or should not be considered an ELF interaction. This results in a hopelessly muddled vision whose main function seems to be what might be called “sanitization,” in the form of a crude syllogism going as follows: “Yes, the spread of English may be imperialistic; but ELF is not English; therefore the use of English in the form of ELF is not imperialistic.”

(16) Let us also point out that the shift to linguistic uniformity is not costless, which is why the compensatory payments mentioned before would need to be paid.

(17) Some authors, however, predict the natural retreat of dominant languages as a result of the development of information technologies and the decentralization that their use makes possible; see Ostler (2010).