IDEAS home Printed from https://ideas.repec.org/a/inm/ormnsc/v68y2022i6p4173-4195.html
   My bibliography  Save this article

“Un”Fair Machine Learning Algorithms

Author

Listed:
  • Runshan Fu

    (Heinz College, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213)

  • Manmohan Aseri

    (Joseph M. Katz Graduate School of Business, University of Pittsburgh, Pittsburgh, Pennsylvania 15260)

  • Param Vir Singh

    (Tepper School of Business, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213)

  • Kannan Srinivasan

    (Tepper School of Business, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213)

Abstract

Ensuring fairness in algorithmic decision making is a crucial policy issue. Current legislation ensures fairness by barring algorithm designers from using demographic information in their decision making. As a result, to be legally compliant, the algorithms need to ensure equal treatment. However, in many cases, ensuring equal treatment leads to disparate impact particularly when there are differences among groups based on demographic classes. In response, several “fair” machine learning (ML) algorithms that require impact parity (e.g., equal opportunity) at the cost of equal treatment have recently been proposed to adjust for the societal inequalities. Advocates of fair ML propose changing the law to allow the use of protected class-specific decision rules. We show that the proposed fair ML algorithms that require impact parity, while conceptually appealing, can make everyone worse off, including the very class they aim to protect. Compared with the current law, which requires treatment parity, the fair ML algorithms, which require impact parity, limit the benefits of a more accurate algorithm for a firm. As a result, profit maximizing firms could underinvest in learning, that is, improving the accuracy of their machine learning algorithms. We show that the investment in learning decreases when misclassification is costly, which is exactly the case when greater accuracy is otherwise desired. Our paper highlights the importance of considering strategic behavior of stake holders when developing and evaluating fair ML algorithms. Overall, our results indicate that fair ML algorithms that require impact parity, if turned into law, may not be able to deliver some of the anticipated benefits.

Suggested Citation

  • Runshan Fu & Manmohan Aseri & Param Vir Singh & Kannan Srinivasan, 2022. "“Un”Fair Machine Learning Algorithms," Management Science, INFORMS, vol. 68(6), pages 4173-4195, June.
  • Handle: RePEc:inm:ormnsc:v:68:y:2022:i:6:p:4173-4195
    DOI: 10.1287/mnsc.2021.4065
    as

    Download full text from publisher

    File URL: http://dx.doi.org/10.1287/mnsc.2021.4065
    Download Restriction: no

    File URL: https://libkey.io/10.1287/mnsc.2021.4065?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    References listed on IDEAS

    as
    1. Shunyuan Zhang & Nitin Mehta & Param Vir Singh & Kannan Srinivasan, 2021. "Frontiers: Can an Artificial Intelligence Algorithm Mitigate Racial Economic Inequality? An Analysis in the Context of Airbnb," Marketing Science, INFORMS, vol. 40(5), pages 813-820, September.
    2. Jon Kleinberg & Sendhil Mullainathan, 2019. "Simplicity Creates Inequity: Implications for Fairness, Stereotypes, and Interpretability," NBER Working Papers 25854, National Bureau of Economic Research, Inc.
    Full references (including those not matched with items on IDEAS)

    Citations

    Citations are extracted by the CitEc Project, subscribe to its RSS feed for this item.
    as


    Cited by:

    1. Xue Wen Tan & Stanley Kok, 2024. "Explainable Risk Classification in Financial Reports," Papers 2405.01881, arXiv.org, revised May 2024.

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. John W. Patty & Elizabeth Maggie Penn, 2022. "Algorithmic Fairness and Statistical Discrimination," Papers 2208.08341, arXiv.org.
    2. Malamud, Semyon & Cieslak, Anna & Schrimpf, Paul, 2021. "Optimal Transport of Information," CEPR Discussion Papers 15859, C.E.P.R. Discussion Papers.
    3. Maria De‐Arteaga & Stefan Feuerriegel & Maytal Saar‐Tsechansky, 2022. "Algorithmic fairness in business analytics: Directions for research and practice," Production and Operations Management, Production and Operations Management Society, vol. 31(10), pages 3749-3770, October.
    4. Heng Xu & Nan Zhang, 2022. "Implications of Data Anonymization on the Statistical Evidence of Disparity," Management Science, INFORMS, vol. 68(4), pages 2600-2618, April.
    5. Semyon Malamud & Andreas Schrimpf, 2021. "Persuasion by Dimension Reduction," Swiss Finance Institute Research Paper Series 21-69, Swiss Finance Institute.
    6. Claire Lazar Reich, 2021. "Affirmative Action vs. Affirmative Information," Papers 2102.10019, arXiv.org, revised Oct 2024.
    7. Ashesh Rambachan & Jon Kleinberg & Sendhil Mullainathan & Jens Ludwig, 2020. "An Economic Approach to Regulating Algorithms," NBER Working Papers 27111, National Bureau of Economic Research, Inc.
    8. Elizabeth Maggie Penn & John W. Patty, 2023. "Algorithms, Incentives, and Democracy," Papers 2307.02319, arXiv.org.
    9. Tengyuan Liang & Pragya Sur, 2020. "A Precise High-Dimensional Asymptotic Theory for Boosting and Minimum-L1-Norm Interpolated Classifiers," Working Papers 2020-152, Becker Friedman Institute for Research In Economics.
    10. Jeffrey D. Shulman & Olivier Toubia & Raena Saddler, 2023. "Editorial: Marketing’s Role in the Evolving Discipline of Product Management," Marketing Science, INFORMS, vol. 42(1), pages 1-5, January.

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:inm:ormnsc:v:68:y:2022:i:6:p:4173-4195. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Chris Asher (email available below). General contact details of provider: https://edirc.repec.org/data/inforea.html .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.