Notes on the Economics of Language and Language Policy

1
Notes on the Economics of Language and Language
Policy

• Part Two

2
Why an economics of language?

• Breton The analytical method of economics
  (rational-choice framework) has found wide
  applicability in recent decades
• political behavior
• marriage family
• schooling
• drug abuse
• etc.

3
What is the economic approach?

• The economic methodology asks two questions
• How do individuals respond to changes in their
  environment (constraints incentives)?
• What are the implications for society as a whole?
• Insights from economics can contribute with (but
  not replace) those from other disciplines.

4
Some applications of the economic approach to
language

• Marschak (1965)
• Why do some (but not all) languages survive?
• How do languages change over time?
• Labor markets, immigration, and discrimination
• Lang (1986, 1993), Robinson (1988), Chiswick
  (1991, 1992), Chiswick and Miller (1995), Lazear
  (1995, 1999)
• Language spread, maintenance, and shift
• Grenier and Vaillancourt (1983), Grin (1990,
  1992, 1996), Pool (1991), Selten and Pool (1991),
  Church and King (1993), Levy (1997), Dalmazzone
  (1999)

5
Overviews and literature surveys

• Breton (1998)
• Vaillaincourt (1982/1983)
• Grin (1996)

6
Why do languages die?

• Its speakers die or geographically disperse
• famine, war (Crystal)
• involuntary assimilation into a dominant
  language (Breton)
• and/or
• Its speakers no longer pass on the language to
  their children (voluntary assimilation).
• This results from individual choices.

7
Why should we care?

• What is the value of a language?
• Two aspects of language
• Private good
• Public good
• Conceptually, what kinds of value or benefit does
  a non-market public good provide?

8
Types of economic value

• Use value
• Enjoyment (utility) from direct use of the good
• Wilderness recreation, enjoyment of art, etc.
• Option value
• The value of preserving the option to use the
  good in the future.

9
Types of economic value (continued)

• Quasi-option value
• The value of information that might become
  available as a result of preserving the good
• New medical uses for plant species
• Existence value
• The value of simply knowing that the good exists,
  independent of any current or expected future use
• Perhaps because we care about the environment, or
  other species, or because we consider them part
  of our national (or humanitys?) treasures or
  identity

10
Conceptually, how can the value of a non-market
good be defined?

• Willingness to pay (WTP) for a good
• The amount of other goods and services an
  individual is willing to give up to use or
  preserve a good.
• Willingness to accept (WTA) compensation for the
  loss of the good.
• Generally, WTA can be much larger than WTP.
• Which measure is correct depends on the nature of
  property rights.
• In the case of wilderness areas (or cultural
  assets) owned by the public, WTA is the
  appropriate measure.

11
Empirically, how can the value of a non-market
good be estimated?

• Revealed-preference methods
• travel costs, land prices, wages
• can provide estimates of WTP or WTA for use by
  observing peoples choices
• Stated-preference methods
• questionnaires, controlled experiments
• necessary to elicit estimates of non-use values,
  since there is no behavioral evidence

12
What is the use value of a language?

• Language is a form of human capital, like any
  other form of knowledge.
• People will invest in learning a second language
  (or transmitting their native language to their
  children, or improving their skills in their
  native language) if the marginal (additional)
  benefit (MB) exceeds the marginal cost (MC).

13
Benefits and costs

• Benefits include
• exchange (of good services, and of ideas)
• enjoyment of culture
• social status
• Costs include
• time
• tuition
• foregone consumption of other goods (affected by
  persons ability, motivation, and
  environment)which provide a measure of WTP for
  language skills

14
Some positive (objective) predictions

• Other things being equal (ceteris paribus),
  younger individuals are more likely than older
  people to invest in learning a second language
  (because longer expected lifetime to recover the
  costs).
• So language policies are more cost-effectively
  targeted at children than at older adults.
• The cost of learning a second language will be
  lower if that language is more similar to the
  native language.
• Benefits of learning (or transmitting) a language
  will be higher, the more valuable the exchanges
  that the individual can undertake with the
  language.

15
Normative implications

• Will individual choices be optimal
  (economically efficient)?
• Probably not, due to externalities
• Network externalities
• Existence (non-use) value
• Diversity externalities
• Quasi-option value
• Hence marginal social benefit (MSB) of language
  skills may exceed the marginal private benefits
  (MPB).

16
Network externalities
But the fifth person creates a new contact for
each of the other four.
In a group of four individuals who can
communicate with each other, each has three
contacts.
E
B
A
A fifth person contemplating joining the network
considers the four contacts he or she would gain.
Hence the marginal social benefit of the fifth
person is the private benefit of his or her new
contacts, plus the external benefit to the other
four members.
C
D
17
Network externalities (continued)

• In general, suppose there are n speakers in a
  language group.
• Then the marginal private benefit (MPB) to the
  n1st speaker from learning the language is the n
  contacts that he or she gains.
• But this creates a new contact for each of the
  other n speakers. So the marginal social benefit
  (MSB) is 2n contacts.
• Since MSB gt MPB, there will be too little
  investment and too few speakers.

18
Lingua francas

• If each of the n individuals speaks a different
  language, it is less costly for each of them to
  learn a second language (a lingua franca) than
  for each of them to learn all of the other n1
  languages.
• Alternatively, language brokers (interpreters
  and translators) might emerge.
• Each individuals optimal choice depends on what
  others do.
• So coordination may improve the situation.
• The result is not necessarily optimal!

19
What will be the result?

• Breton, page 25 The increasing returns that
  derive from network externalities generally give
  rise to multiple equilibriain the case of
  languages, with the possibility that any one of a
  set of languages could become the historical
  equilibrium lingua franca.
• The solution is highly dependent on historical
  accident (path-dependent).
• Why English?
• Britain, then United States, have been dominant
  world powers.

20
Implications

• The use benefits of language may be very
  sensitive to the size of speaker populations.
• Network externalities the public good nature
  of shared medium of communication create a
  strongly self-reinforcing incentive to invest in
  a lingua franca.
• Is there some threshold number of speakers
  below which extinction is assured? If so, on
  what variables might this threshold depend?

21
A biological model
Change in Population
Minimum viable population

0
Population

Carrying capacity
Loss of habitat reduces the carrying capacity and
increases the minimum viable population, making
extinction more likely.
22
Other externalities to members of the speaker
group

• Existence (non-use) value Part of cultural
  identity (a public good)
• To the extent that cultural values are embodied
  in the language, learning of that language will
  have a cultural benefit.
• If this is a minority language confronted with a
  dominant language or lingua franca, then public
  policy may be warranted to preserve and enhance
  the minority language.

23
Externalities to individuals outside the speaker
group

• Diversity has value in and of itself (Crystal)
• Cultural richness as a consumption good
• Quasi-option value What can we learn from
  language? Information about culture
• Crystal Tuyuca (Brazil Colombia) Five
  evidentiary modes
• Navaho Verbs conjugated by aspect rather than
  tense
• Dyirbals mother-in-law dialect and gender
  structure

24
What types of policy intervention can (should?)
be used?

• Incentive-based policies
• Command-and-control approaches

25
Incentive-based policies

• Subsidies to writers, artists, media
• But scale economies in publishing
• Economic development activities to reduce
  outmigration
• Attempts to maintain the habitat

26
Command (regulation)

• Compulsory education
• Cost what is not being learned instead?
• How does this affect learners attitudes?
• Language restrictions on signs, media
• Why? To increase the habitat.
• Zoning property restrictions to limit
  inmigration of non-speakers

27
Questions

• Should policies be targeted at the language, or
  at its environment?
• What role does technology play?
• Has the Internet increased the value of English
  as a lingua franca?
• Or has it reduced the importance of scale
  economies?

28
Some final issues

• Economics of language vs. economy language
  vs. economic linguistics
• Effects of language on income inequality
  (immigration, assimilation)
• How do languages change as they die?
• Machine languages
• Are some languages more efficient than others?