the economics of multilingualism. a case study laura onofri, feem, university of venice...

The Economics of Multilingualism. A Case StudyLaura Onofri, FEEM, University of Venice

"Valuing Cultural Diversity in Cities:

Challenges to Cultural Economics",

Procida (Naples, Italy),

4th of September 2009

2

TABLE OF CONTENTS

Definitions and background

Cultural Economics and Languages

Cultural Economics, Languages and Sustainable Development and Diversity

3

DEFINITIONS AND BACKGROUND

Economics is the science of choices in a world of scarce resources. In economics, choices have both a psychological and a monetary dimension. Every choice implies a trade-off. Every choice generates costs and benefits.

The economist is the scientist that theoretically models and empirically measures costs and benefits of choices, in order to explain, predict and evaluate phrenomena.

4


Cultural Economics is the economic studies of cultural markets and cultural goods and services.

Cultural markets are institutions where demand and supply for cultural goods and services meet.

Cultural markets are particular markets, charachterised by many “market failures”.

Cultural goods and services are both production inputs and final outputs traded in cultural markets

Cultural goods and services are not always exchanged in the market

5


The peculiar features of cultural markets, goods and services refers to the fact they embody a

“cultural added value”.

Sometimes this value is expressed by a “price”, that approximately expresses the value of the good (in terms of production costs, firms mark-up and consumers’ “appreciation” or WTP).

Sometimes this does not happen, therefore, economists needs to assess the value through non-market valuation methods.

6


Cultural Economics is a vary broad field of research and applications, spanning, for instance, from the study of price formation in Old Masters paintings’auctions; to the non-market valuation of public goods; to the study of the ballet dancers market; to the understanding of the life-cycle of cultural exhibitions; to the microeconomic study of different funding instruments to cultural institutions….

In this lectur, I will ONLY focus on the link between cultural and language economics.

7

CULTURAL ECONOMICS AND LANGUAGES

Economics and Cultural Economics can be used to understand, explain, predict and evaluate/measure several aspects of linguistic dimensions and phenomena.

For instance, cultural economics can help us to address the following issues:

Why do people study different languages?

How do individuals choose the foreign language they want to learn?

Why should minority languages be protected?

What is the value that individuals attach to the preservation of a minority language?

Can we define the price/value of a (minority) language?

…….

Cultural Economics, in general, and language economics, in particular are very recent branches of the economic literature. There exist few seminal papers.

8


Cultural Economics, in general, and language economics, in particular are very recent branches of the economic literature. There exist few seminal papers.

9


(V. Ginsburgh, I. Ortuno-Ortin, S. Weber, 2005). Disenfranchisement In Linguistically Diverse Societies: The Case Of The European Union

The authors consider a linguistically diversified society and examine the notion of language disenfran-chisement when some individuals are denied the full access to documents and political process in their native tongues. To calculate the disenfranchisement indices the Dyen percentage cognate matrix of linguistic distances between Indo-European languages is used and is applied to a survey and population data on language proficiency in the European Union.The optimal sets of official languages depends on society's sensitivity against disenfranchisement and comprehensiveness of the chosen language regime. The language situation in the European Union after its last enlargement is discussed.

10


(Fidrmuc J., Ginsburg V., 2007) Languages in the European Union: The Quest for Equality and its Cost

The EU extends equal treatment to all member countries’ official languages. This is costly, especially since many Europeans speak one of the procedural languages, English, French or German, either as a native or a foreign language. The authors develop a simple theoretical model of linguistic-regime choice in a multilingual society and apply the model’s insights to the case of the EU: compute disenfranchisement rates that would result from using only the three procedural languages for all EU business and then proceed to quantify the average cost per person and cost per disenfranchised person associated with providing translations and interpreting into the remaining languages. Both the disenfranchisement rates and costs are shown to vary substantially across the different languages, raising important questions about the economic efficiency of equal treatment for all languages. A proposed, efficient solution would be to decentralize the provision of translations.

11


(Ginsburg V., Gabszewicz, J. Weber S. 2008) Bilingualism and Communicative Benefits

The autors examine patterns of acquiring non-native languages in a model with two languages and two populations with heterogeneous learning skills, where every individual faces a binary choice of learning the foreign language or refraining from doing so. They show that both interior and corner linguistic equilibria can emerge in the framework, and that the fraction of learners of the foreign language is higher in the country with a higher gross cost adjusted communicative benefit. It turns out that this observation is consistent with the data on language proficiency in bilingual countries such as Belgium and Canada. Linguistic equilibria can exhibit insufficient learning which opens the door for government policies that are beneficial for both populations.

12

CULTURAL ECONOMICS AND LANGUAGES: AN APPLICATION

What is in the signs of a main shopping street?

An econometric study of multilingualism

Jasone Cenoz

Durk Gorter

Paulo A.L.D. Nunes

Laura Onofri

13


MOTIVATION:

The study of language signs can contribute to the study of linguistic diversity. It can reflect the different strengths of the languages and the relative integration of their speakers. The language signs are also related to identity and to language policy.

Research question: how different characteristics (type of shop, branch, chain and so on) affect the choice of the language sign?

14


The dataset contains information about the linguistic landscape in Ljouwert (Ljouwert) and San Sebastian (Donostia). In particular, observations were collected in the cities main shopping streets, in order to allow for a certain degree of comparability. The dataset for Ljouwert contains 103 observations (language signs) and related characteristics (typo of sign, language on the sign ans so on); the dataset for San Sebastian contains 104 observations (language signs) and related characteristics.

The corpus of this study includes a complete inventory of the linguistic landscape of just one street in the Basque Country and one street in Friesland. The streets selected were main shopping streets and both had a length of approximately 600 meters. A total of 975 digital pictures of street signs were taken and 207 units were distinguished 207: 104 in Donostia and 103 in Ljouwert. Each establishment but not each sign was considered the unit of analysis (see Cenoz & Gorter 2006). The collected sign convey information about different characteristics of the sign

15


Number Variable Description 1 Type type of the sign 2 Branch Branch indicated by the sign (e.g. street sign, shop, garage,

bar/restaurant; school; bank/ATM; poster: health care; commercial sign; hairdresser; repair shop; official building; estate agent; cinema; hotel; other.

3 Type of shop Sign type of shop (e.g. Clothing; books; food; electronics; furniture, jewellery; perfume; optician; flower; music; suitcases: lottery, other)

4 Chain or Independent

Independent: regional chain; national chain; international chain

5 Number of languages of the sign

Whether the sign contain one or more languages

6 Language of the sign (Lansign)

Whether the sign is written in the official (national), minority languages; international languages a mix of them or in invented languages.

7 First, second and third language on the sign

Whether the (respectively) first, second or third language on the sign is the minority, national, international language or a mix of them.

8 Size Size on multilingual sign 9 Fonts Type of font 10 Information Type of information conveyed in different languages 11 Translation Type of translation (e.g. word to word, free, no translation, an so on) 12 Function Function of the text (informative; symbolic or both) 13 Grammar Grammar use and correctness 14 Lexis Lexis correctness Number of Observations: San Sebastian (Donostia): 104 Ljouwert (Ljouwert): 103

16


After several checks, we estimate by ML the following probit empirical specification:

Prob| LanguageTypej,k = β0 + βjcharachteristicsi + εi,k,j

where, the dependent variable LanguageType represents

the k-th type of language (minority, official, international

or a combination of them) for sign i; β0 is the costant;

βj are estimators for the j-th considered characteristics

contained in the i-th sign and εi,k.j is the error term.

17


Dependent Variable

Independent Variables

ENGLISH (PROBIT

ESTIMATE) .

BASQUE (PROBIT

ESTIMATE)

SPANISH (PROBIT

ESTIMATE)

SPANISH +

BASQUE (PROBIT

ESTIMATE)

BASQUE +

OTHER LANGUAGES

(PROBIT ESTIMATE)

NATIONAL CHAIN 1.316 (0.04)

- - 0.98 (0.15)

-

INDEPENDENT CHAIN

- - 0.32 (0.25)

-0.18 (0.70)

-

INTERNATIONAL CHAIN

1.16 (0.15)

- - - -

SHOP -0.25 (0.09)

-1.47 (0.01)

- 0.68 (0.17)

-

ATM

- - - 0.76 (0.47)

-

COMMERCIAL SIGNS

- - - - 0.063 (0.06)

FASHION SHOP

- - - - 1.08 (0.05)

OTHER - -0.53 (0.37)

- - -

POSTER - - 1.61 (0.006)

- -

BARS AND RESTAURANTS

- -1.10 (0.09)-

-0.64 (0.319)

1.45 (0.01)

1.47 (0.08)

STREET SIGN - 0.17 (0.74)

- 0.34 (0.00)

-

SAME SIZE - - - 1.62 (0.01)

-

SAME INFORMATION 0.48 (0.48)

-

SIZE MAJORITY LANGUAGE BIGGER

- - - - -0.55 (0.00)

CONSTANT -1.78 (0.00)

-0.43 (0.31)

-0.60 (0.00)

-1.56 (0.00)

-0.67 (0.391)

PSEUDO R2 0.16 0.19 0.10 0.32 0.12 LOG-LIKELIHOOD -17.294 -29.51 -59.091 -36.09 -14.7

P-VALUES IN PARENTHESES

18


Dependent Variable Independent Variables

ENGLISH (PROBIT

ESTIMATE)

FRISIAN (PROBIT

ESTIMATE)

DUTCH (PROBIT

ESTIMATE)

DUTCH +

ENGLISH (PROBIT

ESTIMATE)

DUTCH +

OTHER LANGUAGES

(PROBIT ESTIMATES)

INDEPENDENT CHAIN - - - 0.94

(0.02) 0.18

(0.87) REGIONAL CHAIN - - 2.49

(0.03) -0.49 (0.61)

-

NATIONAL CHAIN 1.71 (0.02)

- 0.89 (0.03)

0.60 (0.34)

-

INTERNATIONAL CHAIN

1.77 (0.08)

- - - -

BARS AND RESTAURANTS

- - - -0.39 (0.45)

0.90 (0.48)

FASHION SHOP - - -1.10

(0.06) 0.06

(0.88) 1.19

(0.24)

FURNITURE SHOP - - 0.61 (0.47)

- -

JEWELRY SHOP - - 1.16 (0.03)

- 1.51 (0.26)

SHOP (AS A BRANCH) 5.90 (0.00)

2.14 (0.04)

- - -

OFFICIAL BUILDING - 2.83 (0.06)

- - -

STREET SIGN - 6.12 (0.00)

- 1.85 (0.00)

- -

POSTER

- - - - 0.94 (0.24)

SIZE FOREIGN LANGUAGE BIGGER

- - - 2.44 (0.00)

-0.04 (0.95)

SAME FONT

- - - - 0.07 (0.90)

CONSTANT -7.79

(0.01) -3.93 (0.00)

-0.66 (0.82)

-1.33 (0.00)

-2.39 (0.00)

PSEUDO R2 0.25 0.16 0.17 0.33 0.15 LOG-LIKELIHOOD -15.37 -10.06 -59.05 -39.84 -17.29 P-VALUES IN PARENTHESES

19


English (NL)

English (S)

Dutch

Spanish Frisian

Basque Dutch English

Spanish Basque

Dutch +

Others (NL)

Basque +

Others (S)

International chain; national chain; shop

International chain; national chain

Street sign; national chain regional chain; furniture shop; jewellery shop

Poster; Independent chain

Shop; official building

Street sign

Independent chain; national chain; fashion shops; size foreign language bigger

National chain; Shop; ATM,; Bars, restaurant;street sign; language same size in the sign; same information

Independent chain; bars restaurant, fashion shops, jewellery shop, poster same font

Commercial sign, fashion shop; bars; restaurants

Street sign

Shop National chain; fashion shop

Bars; restaura

nts

- Shops; other ; bars; restau rants

Regional chain; bars restaurant

Independent chain

Size of the foreign language bigger

Size of the majority language bigger

20


Following Piller, the use of English in commercial signs does not seem to be intended to transmit factual information but is used for its connotational value because the audience can recognize that the message is in English and this activates values such as international orientation, future orientation, success, sophistication or fun orientation. This is particular evident in the Netherlands and Ljouwert, where English is very spread and broadly used for commercial purposes, but not in Donostia (the estimated coefficient for shop presents a negative sign).

21


In Donostia, most commercial activities and street signs are bilingual (Basque and Spanish), whilst in Ljouwert, Frisian is used to indicate official buildings and some shops and is less broadly diffused. Frisian is a spontaneous language phenomenon. In our interpretation, Basque and Spanish have to be considered as complementary goods, whose crossed elasticity of demand is positive. On the contrary, Frisian and Dutch are (imperfect) substitutes (since some words overlaps), whose crossed elasticity of demand is negative. Alternatively, Basque and Spanish can be interpreted as two different markets offering different products, whilst Dutch and Frisian can be interpreted as differentiated products produced in the same market.

The results corroborate the important assumption that multilingualism and the choice of the language (even in a street sign) is an individual and a social preference. Understanding individuals linguistic preference structures is preliminary to the target and design of proper linguistic and social policies.

22


Finally, the linguistic and economic motivations of findings are beyond the objectives of this paper. However, in the stream of other economic works on multilingualism, the results corroborate the important assumption that multilingualism and the choice of the language (even in a street sign) is an individual and a social preference. Understanding individuals linguistic preference structures is preliminary to the target and design of proper linguistic and social policies.

23


How do stakeholders value the possibility to have the street sign the way they want (mostly language choice) and how much are they willing to pay for it?

Jasone Cenoz

Durk Gorter

Paulo A.L.D. Nunes

Laura Onofri

24


1) TOURISTS

(587 respondents in San Sebastian and Leeuwarden, different nationalities, gender, age, professions, nationality, proveniencse, spoken languages.)

Question The local authorities are studying the behaviour of the tourists in detail. In order to assess and rank their priorities, we kindly ask you if you were given 100 Euro by the local authorities, how would you allocate this amount of money among the following activities:

____ having the signs the way you want them

____ having clean, public toilets in the city centre

____ having more parking space in the city centre

____ having more written information about the city for tourists

____ having a private guided tour

25


The significance of the estimates is indicated by ***, ** and *, referring to the 1%, 5% and 10% level, respectively; t-values are between brackets.

var3

var14 195

15

75

26


we want to formally test the inverse relationship (price-quantity) typically expressed by the (inverse) demand curve. Therefore, we regress the following simple, linear model: Therefore, we regress the following simple, linear model:

(1) WTPi = α1 + β2biddersi + εi

where the dependent variable is the willingness to pay (bids) and the independent variable is the number of respondents. The model contains a constant and an error term.

We regress the model using the OLS estimation method.

27


Source | SS df MS Number of obs = 5 -------------+------------------------------ F( 1, 3) = 13.26 Model | 1655.56023 1 1655.56023 Prob > F = 0.0357 Residual | 374.439768 3 124.813256 R-squared = 0.8155 -------------+------------------------------ Adj R-squared = 0.7541 Total | 2030.00 4 507.50 Root MSE = 11.172 -m----------------------------------------------------------------------------- WTP | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+--------------------------------------------------------------- quantity | -.5564908 .1527974 -3.64 0.036 -1.04276 -.0702214 constant | 63.70314 7.77647 8.19 0.004 38.95494 88.45134 ------------------------------------------------------------------------------

28


Source | SS df MS Number of obs = 22 -------------+------------------------------ F( 4, 17) = 14.19 Model | 1846.83101 4 461.707751 Prob > F = 0.0000 Residual | 553.168995 17 32.5393526 R-squared = 0.7695 -------------+------------------------------ Adj R-squared = 0.7153 Total | 2400.00 21 114.285714 Root MSE = 5.7043 ------------------------------------------------------------------------------ q101_pr_si~s | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- q102_pr_to~s | -.8525905 .1872529 -4.55 0.000 -1.24766 -.4575215 q103_pr_pa~e | -.9295898 .174268 -5.33 0.000 -1.297263 -.5619163 q104_pr_wr~n | -.9393982 .1414411 -6.64 0.000 -1.237813 -.6409835 q105_pr_tour | -.9877334 .1630312 -6.06 0.000 -1.331699 -.6437676 _cons | 94.19149 10.52232 8.95 0.000 71.99134 116.3916 ------------------------------------------------------------------------------

29

Interpretation

Question 2: If the local government asks for your financial assistance to this program, via an overnight stay tax, how much is your maximum willingness to pay for?

____ having the signs the way you want them

____ having clean, public toilets in the city centre

____ having more parking space in the city centre

____ having more written information about the city for tourists

____ having a private guided tour

The main difference with the above question is that we leave the respondents free to bid their preferred amount of money (open question). The respondent does not chose how to allocate a fixed amount of money received by the government, but chooses to contribute for the good “out of his pockets”.

30


pric

esig

nope

n

quantitysignopen12 104

3

50

31


With respect to the above curve we can highlight some differences:

the total benefits and individual average bids are lower;

the curve is much more skewed, with a very inelastic segment for higher WTP levels

This might imply that the payment vehicle for the SAME public good, really affects the individual willingness to pay.

32


Source | SS df MS Number of obs = 4 -------------+------------------------------ F( 1, 2) = 1.35 Model | 517.49045 1 517.49045 Prob > F = 0.3658 Residual | 769.25955 2 384.629775 R-squared = 0.4022 -------------+------------------------------ Adj R-squared = 0.1033 Total | 1286.75 3 428.916667 Root MSE = 19.612 ------------------------------------------------------------------------------ WTP | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- quantitysi~n | -.2970668 .2561088 -1.16 0.366 -1.399014 .8048803 _cons | 32.03854 13.81564 2.32 0.146 -27.40535 91.48244 ------------------------------------------------------------------------------

33

CULTURAL ECONOMICS AND SUSTAINABLE DEVELOPMENT AND DIVERSITY

Cultural economics, languages and sustainable

developments and diversity is a much broader field of

investigation. The crucial question, in this case, is: how

far can multilingualism and language diversity improve

and/or help sustainable development and sustainable

diversity?

34


The question is very complex and difficult to address.

Some attempts:

Aghion and Howitt, 1992:

endogenous growth theory. Investment in human

capital formation positively increases national GDP.

35


(Ottaviano and Peri, 2004): The Economic Value of Cultural Diversity:

Evidence from U.S. Cities

Authors use data on wages and rents in different US cities to assess

the amenity effects on production and consumption of cultural

diversity as measured by diversity of countries of birth of city

residents. Show that US-born citizens living in metropolitan areas

where the share of foreign-born increased between 1970 and 1990

have experienced a significant average increase in their wage and

in the rental price of their housing. Such finding is economically

significant and robust to omitted variable bias and endogeneity

bias.

.

36


The authors then present a model in which cultural diversity may have

both production and consumption amenity or disamenity effects.

As people and firms are mobile across cities in the long run, the model

implies that the joint results from the wage and rent regressions

are consistent with a dominant production amenity effect of

cultural diversity

37

CULTURAL ECONOMICS AND SUSTAINABLE ECONOMICS AND DIVERSITY

(Ottaviano and Peri, 2004): Cities and Cultures

The authors investigate the existence of wage premium due to cultural

diversity across US cities. Using census data from 1970 to 1990,

find that at the urban level richer diversity is systematically

associated with higher average nominal wages for white US-born

males. They measure cultural diversity in a city using the variety of

languages spoken by city-residents. While the positive correlation

between wages and diversity survives a battery of robustness

checks, it seems to be larger once foreign cultures have been

assimilated. Finally, instrumental variable estimation hints at

causation going from diversity to wages. Comparing real and

nominal wages across cities, they interpret these results as

evidence that diversity enhances productivity.

38

CONCLUSIONS

Cultural economics and language economics are recent, though, promising fields of study, formation and research.

They address many relevant cultural policy points and rely on the scientific method of economics, in order to explain, predict and evaluate choices in the field of multilingualism, linguistic landscapes and sustainable cultural (and linguistic) diversity.

It is a challenge, that has just started…..!

39

Campo S. Maria Formosa30122 Venezia - Italy

tel +39 | 041 | 27 11 400fax +39 | 041 | 27 11 461web http://www.feem.it

Contact:

[email protected]@feem.it

the economics of multilingualism. a case study laura onofri, feem, university of venice...

Documents

languages cultural economics

languages economics

background cultural

language economics

cultural institutions

cultural exhibitions

background economics

cultural added value