Large Language Models (LLMs) are trained on a prodigious corpus of human writing and may reveal human preferences over characteristics of life courses, such as income, longevity, and working conditions. We present OpenAI's GPT-5.4 and a broadly representative sample of Americans with pairs of...

We study the effects of envy (relative consumption concerns), drawing on evidence that preferences typically exhibit present bias. We employ a Ramseytype model with agents who differ in initial capital endowments and account for present-biased envy: agents are naive and care about how their...

I study how the arrival of new private information affects bargaining outcomes. A seller makes offers to a buyer. The buyer is privately informed about her valuation and the seller privately observes her stochastically changing cost of delivering the good. Prices fall gradually at the early...

I study how the arrival of new private information affects bargaining outcomes. A seller makes offers to a buyer. The buyer is privately informed about her valuation, and the seller privately observes her stochastically changing cost of delivering the good. Prices fall gradually at the early...