I lose vs. I earn: Consumer perceived price fairness toward algorithmic (vs. human) price discrimination

Title of the Paper

I lose vs. I earn: Consumer perceived price fairness toward algorithmic (vs. human) price discrimination

Paper Information

• Subject Area: Human-Computer Interaction (HCI), Algorithmic Pricing and Consumer Behavior Research
• Keywords: Algorithmic Pricing, Price Fairness, Price Discrimination, Human-Computer Interaction, Attribution Theory, Consumer Reactions

Research Background and Issues

• Issues or Challenges:

  • Modern businesses increasingly rely on algorithms to set personalized prices, but the impact of such price discrimination on consumers' perceived price fairness has not been fully explored.
  • There is a lack of in-depth investigation into how human versus algorithmic pricing behaviors influence consumer reactions to price discrimination and the psychological mechanisms underlying these effects.

• Significance:

  • Price fairness is a critical factor influencing consumer purchase satisfaction, brand loyalty, and word-of-mouth communication; understanding the impact of algorithmic and human pricing behaviors can help optimize market strategies.
  • Algorithmic pricing is more widely applied than traditional methods, raising concerns about its social fairness and ethical implications.

• Research Motivation and Related Work:

  • Existing research in the HCI field has examined trust and acceptance of algorithmic versus human decision-making, but most studies focus on areas such as healthcare, human resource management, and loan allocation, with limited attention to consumer reactions to algorithmic pricing.
  • Attribution theory suggests that individuals tend to attribute positive outcomes to internal factors (e.g., luck) and negative outcomes to external factors (e.g., blame-shifting). This theory provides a potential explanatory framework for consumer reactions.

Proposed Solution

• Methods and Innovations:

  • A "two-way experimental design" is proposed to quantitatively study consumers' differing reactions to algorithmic versus human pricing and the psychological attribution mechanisms involved.
  • Hypotheses are developed based on attribution theory: consumers' perceived price fairness in response to price discrimination depends on the type of pricing agent and the discrimination outcome.

• Implementation Steps and Key Techniques:

  1. Experimental Design Definition: Two experiments adopt a 2 (pricing agent: algorithm vs. human) × 2 (price discrimination type: unfavorable vs. favorable) design.
  2. Experimental Process:
     • Participants read hypothetical scenarios (e.g., holiday bookings) and experience different types of price discrimination set by algorithms or humans.
     • Measure consumer perceptions of price fairness, attribution mechanisms (e.g., luck attribution and surprise attribution), purchase satisfaction, brand trust, and brand sentiment.
  3. Data Analysis: Use ANOVA variance analysis and PLS structural equation modeling to analyze the relationship between consumer reactions and attribution mechanisms.

Research Findings

• Specific Findings:

  • Consumers perceive unfavorable price discrimination by humans as more unfair, but favorable price discrimination by algorithms as more unfair.
  • Attribution mechanisms explain these differences: consumers tend to attribute unfavorable pricing by humans to intentional behavior, while algorithms are attributed to unintentional behavior; favorable pricing is more likely to be attributed to consumers' own luck, especially when set by humans.

• Advantages and Impact:

  • Achieved a quantitative study of the differential impacts of algorithmic versus human price discrimination, providing a basis for optimizing marketing strategies.
  • Expanded theoretical exploration in the HCI field regarding algorithmic fairness and consumer reactions.

• Experimental or Evaluation Results:

  • Both experiments validated the predicted hypotheses:
    • Experiment 1 demonstrated that consumer reactions to different pricing types are significantly influenced by the interaction between the pricing agent and the type of price discrimination.
    • Experiment 2 confirmed that surprise attribution and luck attribution mediate reactions to unfavorable and favorable price discrimination, respectively.
  • Statistical significance of attribution mechanisms: favorable attribution significantly influences perceived price fairness through the luck mechanism, while unfavorable attribution significantly influences it through the surprise mechanism.

• Limitations and Future Directions:

  • Limitations:
    1. This study did not empirically test the role of algorithm anthropomorphism, though it is hypothesized that such elements may influence consumer attributions.
    2. Individual differences, such as information sensitivity, were not considered in the study's impact on price fairness perceptions.
    3. The focus was solely on personalized pricing, excluding other widely used pricing strategies like dynamic pricing.
  • Future Directions:
    1. Explore how anthropomorphic design of algorithms moderates consumer attributions and perceptions of price discrimination.
    2. Study consumer reactions to dynamic pricing algorithms.
    3. Investigate potential social and privacy risks to refine ethical models for algorithmic pricing.

Managerial and Ethical Implications

• Managerial Implications:

  1. Introducing algorithmic pricing in unfavorable price discrimination may reduce negative consumer reactions, while favorable discrimination is better set by humans.
  2. When designing websites, adjust the anthropomorphic presentation of pricing elements (e.g., simulating human or machine characteristics) based on the type of discrimination.

• Ethical Recommendations:

  1. Develop responsible algorithms: ensure fairness, transparency, and explainability while minimizing implicit data biases.
  2. Enhance consumer information transparency: proactively disclose how algorithms are used while avoiding exacerbating social inequality issues.