Large Language Models Naively Recover Ethnicity from Individual Records

TL;DR

Large language models can accurately infer ethnicity from names across various countries, outperforming traditional methods like BISG, and can be deployed efficiently at scale with fine-tuned small models.

Contribution

This paper demonstrates that large language models can naively recover ethnicity from names with higher accuracy than existing methods, enabling broader and more accurate demographic inference.

Findings

LLMs achieve up to 84.7% accuracy in ethnicity classification.

LLMs outperform BISG, which has 68.2% accuracy on balanced samples.

Fine-tuned small transformer models can surpass BISG accuracy at lower costs.

Abstract

I demonstrate that large language models can infer ethnicity from names with accuracy exceeding that of Bayesian Improved Surname Geocoding (BISG) without additional training data, enabling inference outside the United States and to contextually appropriate classification categories. Using stratified samples from Florida and North Carolina voter files with self-reported race, LLM-based classification achieves up to 84.7% accuracy, outperforming BISG (68.2%) on balanced samples. I test six models including Gemini 3 Flash, GPT-4o, and open-source alternatives such as DeepSeek v3.2 and GLM-4.7. Enabling extended reasoning can improve accuracy by 1-3 percentage points, though effects vary across contexts; including metadata such as party registration reaches 86.7%. LLM classification also reduces the income bias inherent in BISG, where minorities in wealthier neighborhoods are systematically misclassified as White. I further validate using Lebanese voter registration with religious sect (64.3% accuracy), Indian MPs from reserved constituencies (99.2%), and Indian land records with caste classification (74.0%). Aggregate validation across India, Uganda, Nepal, Armenia, Chile, and Costa Rica using original full-count voter rolls demonstrates that the method recovers known population distributions where naming conventions are distinctive. For large-scale applications, small transformer models fine-tuned on LLM labels exceed BISG accuracy while enabling local deployment at no cost.