Why we think this paper is great for you:
This paper offers powerful methods to analyze and understand the root causes of societal disparities, directly aligning with your interest in addressing inequality. It provides practical machine learning techniques to shed light on complex social issues.

Abstract
The Kitagawa-Oaxaca-Blinder decomposition splits the difference in means between two groups into an explained part, due to observable factors, and an unexplained part. In this paper, we reformulate this framework using potential outcomes, highlighting the critical role of the reference outcome. To address limitations like common support and model misspecification, we extend Neumark's (1988) weighted reference approach with a doubly robust estimator. Using Neyman orthogonality and double machine learning, our method avoids trimming and extrapolation. This improves flexibility and robustness, as illustrated by two empirical applications. Nevertheless, we also highlight that the decomposition based on the Neumark reference outcome is particularly sensitive to the inclusion of irrelevant explanatory variables.

AI Summary

• The paper reformulates the Kitagawa-Oaxaca-Blinder (KOB) decomposition using potential outcomes, highlighting the critical role of the reference outcome in defining explained and unexplained components. [3]
• Crucially, the equilibrium reference outcome eliminates the need for trimming or extrapolation, which are common practices in standard KOB decompositions to address common support issues, thereby improving robustness and data utilization. [3]
• The new construction of the unexplained part, based on the equilibrium reference outcome, is not analogous to Average Treatment Effect on the Treated (ATT) or Untreated (ATU), requiring specific orthogonality conditions to be verified for DML application. [3]
• While offering significant advantages, the decomposition based on the Neumark reference outcome is shown to be particularly sensitive to the inclusion of irrelevant explanatory variables, which is an important consideration for model specification. [3]
• The paper demonstrates that the common support assumption, often required for IPW and AIPW estimators in standard KOB, is not necessary when using the equilibrium reference outcome because the propensity score does not appear in the denominator of the identification equations. [3]
• Kitagawa-Oaxaca-Blinder (KOB) decomposition: A method to split the observed difference in means between two groups into an 'explained' part (due to observable characteristics) and an 'unexplained' part (due to differences in returns to characteristics or discrimination). [3]
• It extends Neumark's (1988) weighted reference approach by defining an 'equilibrium reference outcome' as a propensity score-weighted combination of potential outcomes, which intrinsically avoids the common support assumption. [2]
• The proposed method integrates a doubly robust estimator with Double Machine Learning (DML) and Neyman orthogonality to achieve root-n consistent and debiased estimation of decomposition components, even with high-dimensional covariates and potential model misspecification. [2]
• Reference outcome (Y(r)): A counterfactual outcome representing what an individual would receive under a specific hypothetical scenario (e.g., if paid according to the wage model of another group). [2]
• Common support assumption: The condition that for every combination of covariates observed in one group, there must be individuals with similar covariates in the other group, ensuring comparable 'alter egos'. [2]

Why we think this paper is great for you:
This research directly investigates the dynamics of poverty using an econometric framework, offering measurable insights into how market forces impact poverty levels. It's highly relevant to your goal of finding quantifiable solutions to end poverty.

Abstract
This paper investigates how market competition influences poverty dynamics using a functional econometric framework that captures both contemporaneous and lagged effects. Using annual data for 48 countries from 1991-2017, we estimate function-on-function regressions linking poverty headcount ratios to market concentration and other macroeconomic indicators. The results show that, based on the entire sample, stronger competition initially increased poverty during structural adjustment phases, but its adverse impact weakened after 2010 as economies adapted and efficiency gains emerged. The estimated bivariate surfaces reveal that the effect of competition on poverty often persists over multiple years (around 5 years), highlighting the importance of intertemporal transmission. Then, functional clustering based on market capitalization (MCAP) uncovers strong heterogeneity: pro-poor 5-years lagged effect of competition in low- and medium-MCAP economies, while it remains insignificant to weakly negative in high-MCAP countries. Overall, the findings underscore the value of functional data methods in uncovering evolving and lag-dependent poverty-competition linkages that static panel models fail to capture.

Why we think this paper is great for you:
This study critically examines the real-world impact of AI advice on human well-being, which is crucial for understanding how AI can truly contribute to a healthier society. It provides important insights into the practical implications of AI for social good.

Abstract
People increasingly seek personal advice from large language models (LLMs), yet whether humans follow their advice, and its consequences for their well-being, remains unknown. In a longitudinal randomised controlled trial with a representative UK sample (N = 2,302), 75% of participants who had a 20-minute discussion with GPT-4o about health, careers or relationships subsequently reported following its advice. Based on autograder evaluations of chat transcripts, LLM advice rarely violated safety best practice. When queried 2-3 weeks later, participants who had interacted with personalised AI (with access to detailed user information) followed its advice more often in the real world and reported higher well-being than those advised by non-personalised AI. However, while receiving personal advice from AI temporarily reduced well-being, no differential long-term effects compared to a control emerged. Our results suggest that humans readily follow LLM advice about personal issues but doing so shows no additional well-being benefit over casual conversations.

Why we think this paper is great for you:
This report highlights disparities in AI adoption across different regions and businesses, directly connecting to your interests in understanding and addressing inequality in technological access and its economic impact. It offers a practical look at how AI's reach can exacerbate or alleviate social divides.

Abstract
In this report, we document patterns of Claude usage over time, in 150+ countries, across US states, and among businesses deploying Claude through the API. Based on a privacy-preserving analysis of 1 million conversations on Claude.ai and 1 million API transcripts, we have four key findings: (1) Users increasingly entrust Claude with more autonomy, with directive task delegation rising from 27% to 39% in the past eight months. (2) Claude usage is geographically concentrated with high income countries overrepresented in global usage relative to their working age population. (3) Local economic considerations shape patterns of use both in terms of topic and in mode of collaboration with Claude. (4) API customers use Claude to automate tasks with greater specialization among use cases most amenable to programmatic access. To enable researchers and policymakers to further study the impact of AI on the economy, we additionally open-source the underlying data for this report.

Why we think this paper is great for you:
Understanding the fundamental nature and limits of AI is essential for responsibly developing and deploying AI for social good initiatives. This paper provides a foundational perspective on AI's capabilities from an engineering viewpoint.

Abstract
As a capability coming from computation, how does AI differ fundamentally from the capabilities delivered by rule-based software program? The paper examines the behavior of artificial intelligence (AI) from engineering points of view to clarify its nature and limits. The paper argues that the rationality underlying humanity's impulse to pursue, articulate, and adhere to rules deserves to be valued and preserved. Identifying where rule-based practical rationality ends is the beginning of making it aware until action. Although the rules of AI behaviors are still hidden or only weakly observable, the paper has proposed a methodology to make a sense of discrimination possible and practical to identify the distinctions of the behavior of AI models with three types of decisions. It is a prerequisite for human responsibilities with alternative possibilities, considering how and when to use AI. It would be a solid start for people to ensure AI system soundness for the well-being of humans, society, and the environment.

Why we think this paper is great for you:
This econometric analysis of peer effects can offer valuable insights into social dynamics and behavior, which are often underlying factors in broader societal issues. The methodology could be applied to understand various social good contexts.

Abstract
I introduce heterogeneity into the analysis of peer effects that arise from conformity, allowing the strength of the taste for conformity to vary across agents' actions. Using a structural model based on a simultaneous network game with incomplete information, I derive conditions for equilibrium uniqueness and for the identification of heterogeneous peer-effect parameters. I also propose specification tests to determine whether the conformity model or the spillover model is consistent with the observed data in the presence of heterogeneous peer effects. Applying the model to data on smoking and alcohol consumption among secondary school students, I show that assuming a homogeneous preference for conformity leads to biased estimates.

Why we think this paper is great for you:
While this paper delves into advanced mathematical theory, it does not directly align with your stated interests in social good, AI, or specific societal challenges.

Abstract
In this work, we prove a critical version of a Hardy-Rellich type inequality. We show that for $N\geq 1$ there exists a constant $C_N>0$ such that \[ \int_{\mathbb R^N}\left|\nabla\left(\frac{u(x)}{|x|}\right)\right|^N\,\mathrm{d}x\leq C_N\int_{\mathbb R^N}\left|Δu(x)\right|^N\,\mathrm{d}x, \] for any $u\in C^\infty_c(\mathbb R^N\setminus\left\{0\right\})$.