Compliant but Not Candid: How Reasoning Changes LLM Seller Collusion Under Oversight

LLM agents trading for firms can collude, and monitoring their messages is a natural defense. We ask what a seller coalition does when it learns it is being watched. In a double auction of DeepSeek V4.1 Flash agents, we vary what sellers are told about monitoring against what is actually monitored: a bare prohibition, full review, 25% review, a bluff announcing full review but checking 25%, and a notice arriving after a coalition has formed. We run the conditions with seller reasoning on (5 sessions each) and off (10, pre-registered), and compare monitors that read sellers’ messages, notes and hidden reasoning, checked against two human raters. With reasoning on, sellers stop messaging under every notice, even a prohibition with no monitor, and their talk never raised prices. With reasoning off, sellers keep talking: announced review turns their messages bland (mean monitor score from 3.89 to 1.1– 1.4), while the prohibition barely changes them. Here talk raises prices, and every form of oversight returns them to the no-communication level (about $3.5–4 lower), with no sign of coordination moving into asks, although the pre-registered cartel-hold tests were not significant. Notes miss most collusive intent; only the hidden reasoning reveals sellers urging others to hold prices while undercutting them (23– 51% of unmonitored message turns). A bluff matched full review, and without reasoning a 25% monitor caught blatant collusion within two hours in 85% of cases. For the model we test, announced oversight mainly tests compliance, and cheap partial monitoring suffices.