Lang : English
  • English
  • Français
  • हिन्दी
  • Português
  • Español
close× Call Us

Brief 23: Discrimination in Everyday Behavior

In this experiment, Ghanaian men negotiate taxifare well before, directly before, and 1 year after elections to judge whether election proximity affects price discrimination along ethnic and partisan lines.

Link to Full Study

Category: Conflict and Violence

Tags: Elections, Discrimination, Ethnic Cleavages

Date of Publication: Tuesday, May 26, 2015

EGAP Researcher: Kristin Michelitch

Partners: Photo credit: L & K Bosman

Click to Download the Data

Geographical Region: Africa

Research Question:

Does an impending election exacerbate discrimination on ethnic and/or political lines?

Preparer: Seth Ariel Green



Ghana is a West African country with about 26 million citizens and an average GDP per capita of $1858.24 in 2013, according to the World Bank. It is a competitive democracy with two main parties: the New Patriotic Party (NPP) and the National Democratic Congress (NDC). John Atta Mills of the NDC won the 2008 presidential election by less than a single percentage point. The Ashanti and Fanti peoples are commonly associated with the NPP, and the Ewe and Ga peoples with the NDC.

Market price bargaining aka haggling is a common feature of everyday life in Accra and low-income countries more broadly. Taxi rides are typically negotiated on an individual basis.

Michelitch has used these facets of Ghanian life and applied them to an experiment that tries to explain the scholarship has held that elections exacerbate interethnic discrimination in ethnically diverse new democracies. The scholarship she builds upon presents elections as competitions over societal power and resources between political parties, not ethnic groups. Is it possible, however, that elections exacerbate interpartisan rather than interethnic discrimination? Her experiment attempts to understand how ethnic groups are nested within political parties.

Research Design:
Using flyers, Michelitch recruited 18 to 45 year old men in highly trafficked areas to ride along a pre-destined route in the third free taxi that passed them. Subjects were given 3.5 Ghana cedis (US$3.50), and could keep money left over from negotiations. Subjects were given an opening script to repeat in their mother tongues to establish their ethnic and thereby likely political affiliation to taxi drivers. Upon reaching their destinations, they reported the ethnic identity of their taxi driver and how much they paid to take the ride. The experiment was repeated three times: once 6 months before the 2008 election, once directly before, and once a year later.
Michelitch found that coethnics negotiate lower prices from co-ethnic drivers, regardless of the closeness of an election. At election time, however, copartisans negotiate lower prices while non-copartisans were charged significantly more. Interestingly, while an upcoming election increased discrimination between ethnic groups in opposing political parties, it mitigated discrimination between non-coethnics who shared partisan affiliation.
Table I: Discrimination along ethnic and political lines
Election Time
Not Election time
Lowest prices
Lowest prices
Non-Coethnic Copartisan
Low prices
Medium prices
Non-Coethnic Non-Copartisan
High prices
Medium prices
No matter how close it was to the time of the election, Michelitch found that, on average, riders negotiated the lowest prices from coethnic drivers.  However, Michelitch finds that: “at election time and only at election time, the average price riders are able to negotiate from non-coethnic drivers depends critically on partisan affiliation.” Non-coethnics nested within the same political party (copartisans) get lower prices, while non-coethnics associated with the opposing political party are charged more. The difference in prices charged to coethnic partisans and to non-coethnic non-copartisans “at election time is equivalent to a meal of local food”, which is not insignificant in daily life in Accra.
Policy Implications:
  • Elections do not just affect elite competition – the political contest permeates ordinary behavior between citizens typically considered to be outside the realm of politics. If electoral competition trickles down into ordinary economic activities and induces discrimination on partisan lines, it provides a clue as to why elections are often destabilizing at the citizen level.
  • Policy-makers and scholars are often over-focused on ethnic cleavages as a source of societal tension. The root of discrimination may not have anything to do with ethnic differences, but differences related to a different and coinciding group competition. Policy-makers should consider other identities that coincide with ethnic cleavages before labeling tensions as ethnic.
  • Where prices are not fixed, a negative consequence is that discrimination based on group-identity is able to occur. Events affecting group cleavages exacerbate discrimination.