Cheap Talk

Cheap talk is non-binding communication with no formal penalties for dishonesty. Even without enforceability, it can be credible when speakers have aligned incentives or reputational stakes—like a professor whose recommendation carries weight because she gains nothing from lying. Notice this: the same words can be credible or hollow depending on the speaker's incentives.

How Cheap Talk Works

Cheap talk survives because of credibility mechanisms. When a speaker has a reputation to protect, lying about future actions becomes costly even without legal enforcement. A job candidate's claim is credible if it can be checked or later verified, or if exposure would damage their reputation.

The Alignment Principle

Cheap talk becomes most credible when the speaker's interests align with the listener's. A financial advisor recommending index funds instead of high-fee products may signal credibility if she's fee-only (not earning commissions). However, many advisors have conflicts of interest—the alignment depends on her compensation structure, not just the product quality. Contrast this with a used car seller praising a vehicle's reliability; the misalignment raises skepticism regardless of how sincere she sounds.

When Cheap Talk Fails

The classic problem arises when incentives diverge. A CEO announces "no layoffs planned this year" during a crisis. Even if she believes it when speaking, changing circumstances may force her hand. The statement isn't a lie in the moment, but neither is it a guarantee—listeners who treat it as binding information may make decisions they later regret. Political campaign promises face the same credibility gap: candidates often want to deliver, but institutional constraints can prevent follow-through.

Strategic Ambiguity

Sometimes cheap talk is deliberately vague to preserve flexibility. Central bankers use phrases like "accommodative for the foreseeable future" rather than specific dates, allowing adjustment while signaling general intent. This ambiguity isn't deception—it's a recognition that rigid commitments can backfire in complex environments.

Cheap talk emerges from game theory and information economics, particularly the work on signaling games where informed parties communicate with uninformed ones. Crawford and Sobel's seminal 1982 paper established that even when interests aren't perfectly aligned, some information transmission occurs through cheap talk as long as preferences aren't too divergent. The core insight: credible communication requires that the sender's interests not be too divergent from the receiver's.

Key findings include:

• Crawford and Sobel (1982): When sender and receiver preferences are sufficiently similar, cheap talk conveys information; as interests diverge, communication becomes noisier until only babbling remains.[1]

• Farrell and Rabin (1996): Cheap talk can coordinate behavior in games like Battle of the Sexes even when messages are non-binding, because both parties gain from mutual understanding.[2]

• Benabou and Tirole (2011): Focus on how identity, morals, and beliefs function as assets that shape credible communication—individuals may become more reliable or honest when identity, morals, and social norms are salient; the paper is about beliefs as assets, not a general theory of communication credibility.[3]

Cheap talk models assume rational actors who can perfectly calculate incentives—a simplification that breaks down when emotions, cognitive biases, or bounded rationality enter. The framework extends to repeated interactions, but reputation dynamics become more complex as long-term reputation incentives interact with short-term incentives to mislead. Cultural differences matter too: what counts as a credible signal varies across societies, and some contexts (like diplomacy) use layered communication where apparent cheap talk carries hidden meaning through context, timing, or channel selection. Finally, the theory doesn't fully explain why some people lie even when it's against their long-term interest—bounded rationality and psychological factors complicate the pure incentive analysis.