July 3, 2026
When the math starts a market brawl
Markets are competitive if and only if P = NP
Even the title sparked chaos as readers argued whether smarter software means less real competition
TLDR: A new paper claims truly competitive markets depend on some problems staying too hard for computers to solve, and that smarter artificial intelligence could push markets toward quiet price coordination. Commenters were split between “this is huge” and “hold on, the headline is wrong and real-world cartels already exist,” turning the thread into a lively fight over both the theory and the title.
This paper hit the internet like a two-part plot twist: first with a big claim, then with commenters rushing in to say the headline itself was wrong. The original post said markets are competitive if and only if P = NP, but multiple readers immediately sounded the alarm that the paper actually says P != NP. In plain English: people were barely past the title before the comments turned into a fact-checking pile-on. One of the biggest reactions was basically, “Guys, the headline changed the meaning of the whole paper.” Instant drama.
Once that dust-up started, the real debate got even juicier. The author argues that if computers get good enough, companies may be able to quietly keep prices high by spotting cheaters and punishing them—without ever needing an old-school smoke-filled-room conspiracy. That led to the most eyebrow-raising takeaway in the thread: markets might be able to be fast and information-rich or truly competitive, but not both. Some commenters were fascinated and called the policy implications huge. Others were much less impressed, pointing out that real-world cartels and price-fixing already exist, so calling collusion “unstable” felt suspiciously neat compared with messy reality.
And yes, there was also some dry internet comedy: users gleefully noted that the author appears to be building a whole cinematic universe of papers, including an older one claiming markets are efficient if and only if P = NP. For commenters, that was half profound theory, half “this sequel dropped right on schedule.”
Key Points
- •The paper argues that competitive market outcomes require computational intractability.
- •It claims that if P = NP, firms can efficiently detect deviations from collusive agreements and sustain collusion as an equilibrium.
- •It claims that if P != NP, collusion detection is computationally infeasible in markets with a natural instance-hardness condition on demand structure.
- •The article combines this result with Maymin (2011), which it says proved that market efficiency requires P = NP.
- •The abstract argues that artificial intelligence increases firms' computational capabilities and may be driving the emergence of algorithmic collusion without explicit coordination.