race and gender discrimination in bargaining for a new car

Race and Gender Discrimination in Bargaining for a New Car

Ian Ayres

Peter Siegelman

The Method

4 Groups AnalyzedWhite MalesWhite FemalesBlack MalesBlack Females

Audit TechniquePairs (one always white male)

The Method

38 testers 306 cars 153 dealerships in Chicago Nine car models Conducted at Off-peak hours

The Method

RandomnessDealershipsTesters to DealershipsComposition of PairsFirst to Enter

The Method – Bargaining

Each tester followed a “uniform bargaining script that instructed them to focus quickly on one particular car and start negotiating over it. At the beginning of the bargaining, testers told dealers that they ould provide their own financing.”

The Method – Bargaining Strategies Initial offer

Counter offer equal to the dealer’s marginal cost for the car

“split-the-difference” strategy “fixed-concession” strategy

Controls and Uniformity

Age: all testers were between 28 and 32 years old

Education: All testers had 3-4 years of postsecondary education

Attractiveness: All testers were subjectively chosen to have average attractiveness.***

Controls and Uniformity

Similar signs of economic classAll testers wore similar “yuppie” sportswear

and drove to the dealership in similar rented cars

Testers had a long list of contingent responses to the questions they were likely to encounter (i.e. career or address)

Controls and Uniformity

Two days of trainingMock negotiationsMemorized bargaining script

Testers did not know the circumstances of the study

Trends

The differences in prices quoted are robust to a variety of alternative specifications.Fixed vs. Random Effects Individual-Tester EffectsAttempted acceptances versus Refusals

Trends

Nonparametric Tests for Race and Gender

Trend Conclusions

1) Both final and initial offers display large and significant differences in outcomes by race and gender.

2) Results are robust. The magnitude and significance of the race and gender effects under various alternative specifications, combined with the insignificance of the individual-tester effects, reinforce conclusions.

Sources of Discrimination

Animus based discrimination Statistical Discrimination

Animus

Owner animus Employee Animus Customer Animus

Evidence argues against animus-based theories as primary explanation

Statistical Theories

Search CostsLikelihood of owning car when they already

have oneQuestions about owning a car, visiting other

dealershipsBlacks in the suburbs

Statistical Theories

Consumer InformationNegotiable Sticker PriceOffers given at Sticker Price

Statistical Theories

Bargaining CostsBuyer’s aversion to conducting negotiations

Conclusion

“Chicago car dealers offered black and female testers significantly higher prices than the white males with whom they were paired, even though all testers used identical bargaining strategies”