Kanchan Chandra, Why Ethnic Parties Succeed: Patronage and head counts in India (2004), Chapter 2.

‘Main Argument: This chapter proposes a connection between limited information and the use of ethnic cues to identify and distinguish between individuals. The authors contends that limited information settings are sufficient to produce a tendency toward ethnic categorisation, but not necessary.

 

Key Definitions:

Limited Information Situation: A decision-making situation in which observers are called upon to identify and distinguish between individuals under server informational constraints.

== Notes ==

 

Themes of the Book:

  1. People are motivated by either material or “psychic” (e.g. esteem needs, a la Horowitz) goods, or a combination of both
  2. People are instrumentally rational; they vote for the party that will maximize these goods, not just to express an ethnicity
  3. An ethnic group’s elites (urban, educated, better off) desire state employment or political office as the best prospect of material advancement (this theory applies to clientelistic democracies)

– This requires that they be close to a political patron

  1. Voters in patronage democracies have information constraints that lead voters and politicians to favour co-ethnics

– Voters: Will support any party that puts elites from their ethnic group in power. Count heads; whichever party has the most elites from your ethnic group is the one that you (sincerely) prefer

– Elites: Parties with internally competitive advancement do better at attracting an ethnic group’s elites

5.Voters are strategic. They use elite head counts (4a, above) to form sincere preferences, then they use headcounts of their co-ethnic voters to adjust these preferences according to strategic voting

 

Chapter 2: Limited Information and Ethnic Categorisation

– This chapter proposes a connection between limited information and the use of ethnic cues to identify and distinguish between individuals

– Limited information settings are sufficient to produce a tendency toward ethnic categorisation, but not necessary

– The absence of costless data on ethnic identities, combined with the scarcity of costless data on non-ethnic identities, biases individuals toward schemes of ethnic categorisation in limited information settings

– People tend to use visible identity cues to judge people in political settings

– Popkin was one of the first to recognise this; Posner also recognises this factor, however, it was largely ignored by some of the mainstream voting literature (ie. Downs) [36]

 

Costless information: name, features, speech, manner of dress

Costly information: education, income, ideology, biography, references

 

*Most sources of costless information can be misleading (i.e. names no longer hold information about class, profession, income because population has grown and integrated so much)

 

– Precision of ethnic categorisation is increased when multiple data sources are introduced [43]

– Different observers would code people differently based on their own personal experiences

– When coding based on external features, considerable uncertainty remains (too much opportunity for error)

– The categories in which a coder may place an individual may have no bearing on the identity groups that the individual actually associates with (i.e. these categories may not be readily available  if the individual associates with “scholar” and not “Asian”)

 

Using Limited Information in Electoral Contexts:

– These types of codings are used y voters in elections, particularly when the candidate or the entire electoral context is new [46]

– Competitive environments are more likely to resemble limited information settings

Leave a Reply

Your email address will not be published. Required fields are marked *