Gender Bias and Property Taxes (2412.12610v2)

Abstract: Gender bias distorts the economic behavior and outcomes of women and households. We investigate gender biases in property taxes. We analyze records of more than 100,000 property tax appeal hearings and more than 2.7 years of associated audio recordings, considering how panelist and appellant genders associate with hearing outcomes. We first observe that female appellants fare systematically worse than male appellants in their hearings. Second, we show that, whereas male appellants' hearing outcomes do not vary meaningfully with the gender composition of the panel they face, those of female appellants' do, such that female appellants obtain systematically lesser (greater) reductions to their home values when facing female (male) panelists. Employing a multi-modal LLM (M-LLM), we next construct measures of participant behavior and tone from hearing audio recordings. We observe markedly different behaviors between male and female appellants and, in the case of male appellants, we find that these differences also depend on the gender of the panelists they face (e.g., male appellants appear to behave systematically more aggressively towards female panelists). In contrast, the behavior of female appellants remains relatively constant, regardless of their panel's gender. Finally, we show that female appellants continue to fare worse in front of female panels, even when we condition upon the appelant's in-hearing behavior and tone. Our results are thus consistent with the idea that gender biases are driven, at least in part, by unvoiced beliefs and perceptions on the part of ARB panelists. Our study documents the presence of gender biases in property appraisal appeal hearings and highlights the potential value of generative AI for analyzing large-scale, unstructured administrative data.

• The paper analyzes 114,515 property tax appeal hearings in Harris County, Texas, finding that female appellants are less likely to secure favorable outcomes compared to males.
• Outcomes for female appellants are negatively impacted by panel composition, particularly when facing all-female Appraisal Review Boards, highlighting complex gender dynamics.
• Behavioral analysis using AI reveals that while male appellants' aggressive tactics don't hurt them, gender biases persist for females even when controlling for behavior, suggesting implicit factors are at play.

Gender Bias and Property Taxes: An Analysis of Appraisal Review Board Hearings

This paper, authored by Gordon Burtch and Alejandro Zentner, explores the intricate dynamics of gender bias within the context of property tax appeals, specifically focusing on hearings conducted by Appraisal Review Boards (ARB) in Harris County, Texas. Through a comprehensive analysis of 114,515 hearing records and associated audio data, the research uncovers systematic gender disparities, particularly the pronounced disadvantage faced by female appellants in securing favorable outcomes.

Key Findings

1. Gender-Based Disparities: The paper reveals that female appellants consistently receive less favorable outcomes compared to their male counterparts. The probability of female homeowners obtaining reductions in their assessed property values is notably lower, demonstrating a systemic bias that cannot be attributed solely to the qualities of the cases presented.
2. Impact of Panel Composition: The research finds that outcomes for female appellants are significantly influenced by the gender composition of the ARB panels. Female appellants are approximately 4.2 percentage points less likely to achieve any reduction in their home value when facing an all-female panel compared to an all-male one. This negative gender concordance effect underscores the complexity of identity dynamics in evaluative processes.
3. Behavioral Dynamics: Leveraging advances in multimodal AI technology, the authors analyze over 2.7 years of audio data to assess behavioral differences during hearings. Interestingly, while male appellants often exhibit more aggressive behaviors, such as raising their voices and interrupting, these actions do not negatively impact their outcomes. Conversely, female appellants' behaviors and tones show minimal variation with panelist gender, suggesting that unvoiced perceptions may primarily drive the observed biases.
4. Conditional Effects: The persistence of outcome disparities even after controlling for appellant behavior strongly suggests that the documented gender biases are driven by deeper, implicit factors within the decision-making processes of ARB panelists.

Theoretical and Practical Implications

The findings of this paper hold significant implications both theoretically and practically. From a theoretical standpoint, the research contributes to the literature on gender biases by highlighting a context where gender (dis)concordance effects manifest contrary to commonly expected patterns. By showing that female panelists adjudicate more critically against female appellants, this paper challenges the notion that demographic concordance inherently leads to favorable outcomes.

Practically, the paper emphasizes the urgent need for policy interventions aimed at enhancing fairness in administrative processes. Given the essential role property taxes play in funding local services, gender biases in appraisal hearings may translate into broader social inequities. Policymakers are encouraged to consider trialing structural modifications, such as varying the gender composition of panels or implementing bias awareness training to mitigate these biases.

Methodological Advancements and Future Directions

The utilization of a multi-modal LLM (M-LLM) for processing extensive audio recordings marks a significant methodological advancement. This approach demonstrates the potential of generative AI in effectively analyzing unstructured administrative data, allowing researchers to uncover nuanced insights that would otherwise be infeasible with traditional manual coding methods. The paper sets a precedent for future research to harness AI technologies in investigating socio-legal processes across different domains and jurisdictions.

Moving forward, further explorations are warranted to assess the generalizability of these findings beyond Harris County. Comparative analyses involving multiple counties and states could illuminate the universality of observed biases and their variations across different regulatory frameworks and cultural contexts. Additionally, future work could focus on experimental interventions within the appraisal process to evaluate the efficacy of proposed solutions in reducing gender bias.

In conclusion, this paper provides a robust and detailed account of gender disparities in property tax hearings, presenting critical insights into the pervasive influence of gender bias in administrative operations. By integrating cutting-edge AI tools into socio-economic research, the authors offer novel pathways for addressing inequities in institutional practices.