# (Machine) Learning What Policies Value

arXiv: https://arxiv.org/abs/2206.00727 · HTML: https://mlsanity.com/paper/2206.00727/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 43 references that are in the map.
- First posted on arXiv 1 Jun 2022.

## Builds on (most cited first, 7 of 7)

- [Estimation and Inference of Heterogeneous Treatment Effects using Random Forests](https://mlsanity.com/paper/1510.04342/) (2015)
- [Delayed Impact of Fair Machine Learning](https://mlsanity.com/paper/1803.04383/) (2018)
- [Runaway Feedback Loops in Predictive Policing](https://mlsanity.com/paper/1706.09847/) (2017)
- [From Fair Decision Making to Social Equality](https://mlsanity.com/paper/1812.02952/) (2018)
- [Balancing Competing Objectives with Noisy Data: Score-Based Classifiers for Welfare-Aware Machine Learning](https://mlsanity.com/paper/2003.06740/) (2020)
- [Welfare and Distributional Impacts of Fair Classification](https://mlsanity.com/paper/1807.01134/) (2018)
- [Evaluating Ex Ante Counterfactual Predictions Using Ex Post Causal Inference](https://mlsanity.com/paper/1806.07016/) (2018)

## Built on by (most cited first, 4 of 4)

- [Welfare as a Guiding Principle for Machine Learning -- From Compass, to Lens, to Roadmap](https://mlsanity.com/paper/2502.11981/) (2025)
- [Perceptions and Detection of AI Use in Manuscript Preparation for Academic Journals](https://mlsanity.com/paper/2311.14720/) (2023)
- [Reduced-Rank Multi-objective Policy Learning and Optimization](https://mlsanity.com/paper/2404.18490/) (2024)
- [What Would it Cost to End Extreme Poverty?](https://mlsanity.com/paper/2609.02013/) (2026)

## Abstract (arXiv)

When a policy prioritizes one person over another, is it because they benefit more, or because they are preferred? This paper develops a method to uncover the values consistent with observed allocation decisions. We use machine learning methods to estimate how much each individual benefits from an intervention, and then reconcile its allocation with (i) the welfare weights assigned to different people; (ii) heterogeneous treatment effects of the intervention; and (iii) weights on different outcomes. We demonstrate this approach by analyzing Mexico's PROGRESA anti-poverty program. The analysis reveals that while the program prioritized certain subgroups -- such as indigenous households -- the fact that those groups benefited more implies that they were in fact assigned a lower welfare weight. The PROGRESA case illustrates how the method makes it possible to audit existing policies, and to design future policies that better align with values.