Research Article

The distributional preferences of an elite

See allHide authors and affiliations

Science  18 Sep 2015:
Vol. 349, Issue 6254, aab0096
DOI: 10.1126/science.aab0096
  • Classifying subjects’ distributional preferences.

    We classify subjects as either fair-minded, intermediate, or selfish and as either equality-focused or efficiency-focused. The bars show the fraction of subjects in each category of self-interest in the elite YLS, UCB (the intermediate elite), and relatively less elite ALP samples. Each bar is then split into equality-focused and efficiency-focused subgroups, denoted by blue and gray, respectively.

  • Fig. 1 CCEIs.

    (A and B) Histograms of the CCEI in (A) the YLS and ALP samples and (B) the ALP elite versus nonelite samples. CCEIs closer to 1 mean the data are closer to perfect consistency with GARP and hence to perfect consistency with utility maximization.

  • Fig. 2 Classifying subjects’ distributional preferences.

    We classify subjects as either fair-minded (Embedded Image <0.95), or selfish (Embedded Image > 0.95) and as either equality-focused (Embedded Image < 0) or efficiency-focused (Embedded Image > 0). The bars show the fraction of subjects in each category of self-interest in the YLS and ALP samples. Each bar is then split into equality-focused and efficiency-focused subgroups, denoted by blue and gray, respectively.

  • Fig. 3 Estimated Embedded Image parameters.

    (A and B) Histograms of the Embedded Imageestimates in (A) the YLS and ALP samples and (B) the ALP elite versus nonelite samples. Embedded Image indexes fair-mindedness: the relative utility weight placed on one’s own payoff vis-à-vis the payoff to other.

  • Fig. 4 Estimated Embedded Image parameters.

    (A and B) Histograms of the Embedded Imageestimates in (A) the YLS and ALP samples and (B) the ALP elite versus nonelite samples. Embedded Image indexes equality-efficiency tradeoffs; Embedded Image values closer to 1 indicate greater efficiency focus.

  • Table 1 Summary statistics on subjects in YLS and ALP samples.
    Subject pool
    YLS subjectsALP subjects
    Age25.4031.23
    Female0.4660.653
    Born in the United States0.7820.906
    Non-Hispanic white0.6320.545
    Completed college10.327
    Observations208309
  • Table 2 Regressions of estimated CES parameters, by subject pool.

    Standard errors are in parentheses. ***, significance at the 99% level; **, significance at the 95% level; *, significance at the 90% level.

    Quantile regressions
    SpecificationTobit25th percentile50th percentile75th percentileProbit
    Dependent variableEmbedded ImageEmbedded ImageEmbedded ImageEmbedded ImageEmbedded Image
    Without controls
    YLS student0.120***1.075***0.693***0.405***0.831***
    (0.017)(0.217)(0.080)(0.049)(0.122)
    Constant0.668***–0.867***0.0050.420***0.004
    (0.011)(0.137)(0.051)(0.031)(0.071)
    Observations517517517517517
    Including controls for age, gender, and education level
    YLS student0.084***0.5520.357***0.260***0.440**
    (0.027)(0.476)(0.120)(0.077)(0.183)
    Constant0.660***–0.0510.260**0.604***0.497***
    (0.029)(0.480)(0.121)(0.078)(0.183)
    Observations514514514514514
  • Table 3 Ordered logit estimation of YLS subjects’ career choices.

    Standard errors in parentheses. ***, significance at the 99% level; **, significance at the 95% level; *, significance at the 90% level. Dependent variable is equal to 1 for subjects who work in the nonprofit sector, equal to 2 for subjects who work in academia or government, and equal to 3 for subjects who work in the corporate sector. Controls are for age, gender, and year of experimental session.

    Dependent variable: post-YLS career category
    Without controls
    Above median Embedded Image1.043***
    (0.364)
    Decile of estimated Embedded Image0.157**
    (0.068)
    Observations120120
    With controls
    Above median Embedded Image1.035***
    (0.374)
    Decile of estimated Embedded Image0.164**
    (0.076)
    Observations118118

Supplementary Materials

  • The distributional preferences of an elite

    Raymond Fisman, Pamela Jakiela, Shachar Kariv, Daniel Markovits

    Materials/Methods, Supplementary Text, Tables, Figures, and/or References

    Download Supplement
    • Materials and Methods
    • Figs. S1 to S3
    • Tables S1 and S2

Stay Connected to Science

Navigate This Article