Strategy — Negotiation

Game Theory in Supplier Negotiations: Why Both Sides Lose When They Should Not

Mutual defection is the rational outcome in buyer-supplier games — yet both sides walk away worse. Restructuring the payoff matrix through game-theoretic sourcing recovers 100-200 basis points that standard negotiation leaves on the table.

100–200 bps

Incremental savings from game-theoretic design

Like finding $6–10M extra on a $1B spend base

$6–10M

On a $1 billion annual spend

Enough to fund a whole procurement transformation

50%

Of spend base where game theory applies

Half your suppliers are playing a game you can redesign

The Prisoner's Dilemma in Procurement

Common

Buyer asks for best price. Supplier embeds 15% margin. Both settle at a ~4% reduction — stable, predictable, and 100–200 bps below what the structure could produce.

Structure rewards opacity

Correct

Design sourcing as a repeated game with binding rules. Parallel negotiations create a prisoner's dilemma between suppliers — each fears the other will undercut, driving truthful bidding.

Structure rewards transparency

How to Apply It

Classify your games first. Repeated, indefinite-relationship categories need tit-for-tat cooperation. One-shot categories need auction mechanisms with binding selection rules.

Run parallel negotiations. Put 2–3 suppliers in competition and iteratively revise a floor price. Each fears losing the contract — exactly the dynamic that drives price discovery.

Audit credible commitment. If suppliers suspected transparency would be exploited in your last three sourcing events, the equilibrium was opacity — and that is what you received.

Where the Model Breaks Down

Risk

Finite games with known endpoints predict mutual defection. If a supplier knows the contract ends in December regardless of performance, defection is rational in every round leading up to it. Open-ended, indefinite relationships are essential for sustaining cooperation.

Jargon Decoder

Nash equilibrium The stable outcome where no player can improve by changing strategy alone — like two people stuck in a silent standoff because moving first loses.

Payoff matrix A simple table showing what each side gains or loses from every combination of choices — like comparing two restaurant menus side by side.

Tit-for-tat Cooperate on move one, then copy your opponent's last move. Winning strategy in repeated games — be nice, retaliate when crossed, forgive quickly.

Basis points (bps) 1 bp = 0.01%. 100 bps = 1%. Think of it as pennies per $100 — a 200 bps saving is like getting 2% more value from every dollar spent.

Iterated game A game played repeatedly where players remember past moves, unlike a one-shot interaction. Repetition is what makes cooperation possible.

Backward induction Reasoning from the end of a game backward — if the last move is "everyone defects," that logic ripples back to the first move unless the endpoint is unknown.

Sources: Genpact, Stanford Encyclopedia of Philosophy, Axelrod (The Evolution of Cooperation), Kloepfel Consulting, Academia.edu (2025 study), Alvarez & Marsal, Arkestro

Rzzro

Procurement, quantified.