Achieving Fairness and Accuracy in Regressive Property Taxation

Ozan Candogan,Feiyu Han,Haihao Lu

2023-12-11

Abstract:Regressivity in property taxation, or the disproportionate overassessment of lower-valued properties compared to higher-valued ones, results in an unfair taxation burden for Americans living in poverty. To address regressivity and enhance both the accuracy and fairness of property assessments, we introduce a scalable property valuation model called the $K$-segment model. Our study formulates a mathematical framework for the $K$-segment model, which divides a single model into $K$ segments and employs submodels for each segment. Smoothing methods are incorporated to balance and smooth the multiple submodels within the overall model. To assess the fairness of our proposed model, we introduce two innovative fairness measures for property evaluation and taxation, focusing on group-level fairness and extreme sales price portions where unfairness typically arises. Compared to the model employed currently in practice, our study demonstrates that the $K$-segment model effectively improves fairness based on the proposed measures. Furthermore, we investigate the accuracy--fairness trade-off in property assessments and illustrate how the $K$-segment model balances high accuracy with fairness for all properties. Our work uncovers the practical impacts of the $K$-segment models in addressing regressivity in property taxation, offering a tangible solution for policymakers and property owners. By implementing this model, we pave the way for a fairer taxation system, ensuring a more equitable distribution of tax burdens.

Optimization and Control

What problem does this paper attempt to address?

The problem this paper attempts to address is the unfairness and inaccuracy in property tax assessments, particularly in cases where low-value properties are overvalued and high-value properties are undervalued, leading to a disproportionate tax burden on low-income families. Specifically, the paper focuses on how to mitigate this regressivity issue by improving machine learning algorithms, thereby enhancing the fairness and accuracy of property tax assessments.

### Main Research Questions:
1. **Reducing the Regressivity of Property Tax Assessments**: Regressivity refers to the phenomenon where low-value properties are overvalued and high-value properties are undervalued, resulting in low-income families bearing a higher effective tax rate.
2. **Improving the Fairness and Accuracy of Assessments**: By introducing new assessment models, ensure that the assessed values of all properties are closer to their actual market values, while ensuring the assessment process is fair for properties of different values.

### Solution:
To address the above issues, the paper proposes a new method called the K-segment model. The main features of this model include:
- **Data Segmentation**: Dividing the data into K segments, with each segment using an independent sub-model for assessment.
- **Smoothing Techniques**: Introducing smoothing techniques to eliminate discontinuities at segment boundaries, ensuring smooth transitions in assessment results.
- **Fairness Metrics**: Proposing two new fairness metrics to evaluate the fairness of the model, particularly for the assessment of low-value and high-value properties.

### Specific Contributions:
1. **New Fairness Metrics**: Proposing two new fairness metrics, Group Fairness and Deviation-Weighted Fairness, to quantify the unfairness of different assessment models.
2. **K-segment Model**: Proposing a new K-segment model that improves the accuracy and fairness of assessments by dividing the data into multiple segments and applying independent sub-models within each segment.
3. **Smoothing Techniques**: Introducing various smoothing techniques, such as quantile-based smoothing and score-based smoothing, to eliminate discontinuities at segment boundaries and ensure smooth transitions in assessment results.

### Experimental Results:
The paper validates the effectiveness of the K-segment model through numerical experiments. The experimental results show that, compared to existing assessment models, the K-segment model not only significantly improves assessment accuracy (higher R² values) but also effectively reduces regressivity and enhances fairness.

### Conclusion:
By introducing the K-segment model and its smoothing techniques, the paper provides a practical solution that can effectively mitigate the regressivity issue in property tax assessments and improve the fairness and accuracy of assessments. This approach is not only of significant practical importance to policymakers and property owners but also offers new insights for research in the field of fair regression.