Hotwife-Bull Bargaining: A Nash Solution

Welcome back. In our last lesson, we used the Shapley value to determine a fair way to distribute the value created by a cooperative team. We addressed the "how to divide the pie" problem, assuming cooperation was a given. But how do rational individuals arrive at a specific cooperative agreement in the first place, especially when their interests conflict?

This brings us to the focus of today's lesson: the negotiation process itself. We will explore how two parties can move from a set of possibilities to a single, mutually acceptable outcome. Our objective is to formulate a cooperative bargaining model, specifically the Nash Bargaining Solution, for the hotwife and bull to negotiate the terms of their encounters and achieve a mutually beneficial outcome. This powerful framework, developed by John Nash, uses a set of intuitive "fairness" axioms to identify a unique, optimal agreement.

1. Defining the Bargaining Problem

Before finding a solution, we must first structure the problem. A two-person bargaining problem consists of two key components: the feasible set of outcomes and the disagreement point.

Let's watch a short video that clearly defines this framework.

(AGT5E1) [Game Theory] Cooperative (Nash) Bargaining Problem: The Framework

This video from Selcuk Ozyurt's game theory series provides a clear introduction to the components of a cooperative bargaining problem.

Please watch the segment from the beginning, which introduces the three core elements of a bargaining problem. Focus on: The set of players (negotiators). The set of feasible payoff profiles, S (payoff profile). The disagreement point, d (disagreement point).

To apply this to our scenario, let's define the components for the hotwife (H) and bull (B) negotiating the terms of their encounters:

Players: The Hotwife and the Bull.
The Feasible Set (S): This is the set of all possible combinations of satisfaction (utility) that H and B can achieve together. For example, if they are negotiating how to divide 10 hours of shared time per month, each potential division of that time corresponds to a utility pair $(u_{H}, u_{B})$ . The collection of all such pairs forms the feasible set $S$ . We assume this set is convex, meaning that lotteries between feasible outcomes are also feasible.
The Disagreement Point (d): This is the utility outcome for each person if they fail to reach an agreement. It is their "Best Alternative to a Negotiated Agreement" (BATNA). For now, let's assume their BATNA is to have no encounters, yielding a utility of zero for both. So, $d = (d_{H}, d_{B}) = (0, 0)$ . This point serves as the baseline from which any agreement must be an improvement.

A bargaining problem is thus formally described by the pair $(S, d)$ . The goal is to find a function that selects a single, "fair" point from $S$ .

2. The Axioms of a Fair Solution

Instead of modeling a back-and-forth negotiation process, John Nash took an axiomatic approach. He proposed that any reasonable solution to a bargaining problem should satisfy a handful of desirable properties.

The Two Person Bargaining Problem - Game Theory lab

This document from Prof. Y. Narahari provides a formal treatment of the Nash Bargaining solution. We will focus on the axioms that define the solution.

Please read section 3, The Axioms of Nash (pages 4-7). As you read, focus on understanding the intuition behind each of the five axioms: Strong Efficiency (or Pareto Optimality): The agreement should not leave any "value on the table." It should be impossible to improve one person's utility without harming the other's. Individual Rationality: The agreement must give each player at least as much utility as they would get from the disagreement point. No one should be forced into a deal that's worse than walking away. Scale Covariance (or Invariance): The solution shouldn't depend on the arbitrary units used to measure utility. If we change a player's utility scale from 1-10 to 1-100, the underlying outcome should remain the same. Independence of Irrelevant Alternatives (IIA): If the ideal agreement is available in a smaller set of options, it should still be the chosen agreement. Removing unchosen "irrelevant" alternatives shouldn't change the outcome. Symmetry: If the players are in perfectly symmetric positions (i.e., the feasible set is symmetric and their disagreement payoffs are equal), the solution should give them equal payoffs.

These axioms are not just abstract principles. They are powerful constraints. Nash proved that there is one, and only one, solution that satisfies all of them.

3. The Nash Bargaining Solution

The unique solution that fulfills all of Nash's axioms is the utility profile $(u_{H}^{*}, u_{B}^{*})$ that maximizes the Nash Product:

(u_{H}, u_{B}) \in S max (u_{H} - d_{H}) (u_{B} - d_{B})

This objective function represents the product of the gains from cooperation for each player. By maximizing this product, the solution finds a point that balances the players' competing desires for a larger share of the surplus generated by their agreement. It is inherently a cooperative concept, as the product is zero if either player fails to gain anything over their disagreement point.

4. Application: Negotiating an Intimacy Arrangement

Let's apply this to the hotwife-bull dynamic. Suppose they are negotiating how to divide a "budget" of $T = 12$ hours of shared time per month. Their disagreement point is $d = (0, 0)$ . We will analyze two cases based on their utility functions.

The following video walks through two very similar examples. We will follow its methodology precisely.

(AGT5E13) [Game Theory] Solving for Nash Bargaining Outcome: Two Examples

Selcuk Ozyurt's video "Solving for Nash Bargaining Outcome" provides an excellent, step-by-step guide to calculating the solution in both symmetric and asymmetric cases.

We will apply the logic demonstrated in this video. You can watch it to see the calculations for a generic "cake-splitting" problem, which we will adapt. The key parts are the setup (setup), the symmetric case (symmetric), and the asymmetric case (asymmetric).

Case 1: Symmetric, Risk-Neutral Partners

Assume both the hotwife and bull are risk-neutral, meaning their satisfaction is directly proportional to the time they get. Let $t_{H}$ be the time allocated to activities she prioritizes, and $t_{B}$ be the time for his.

Constraint: $t_{H} + t_{B} = 12$
Utility Functions: $u_{H} = t_{H}$ and $u_{B} = t_{B}$
Feasible Set Boundary: From the constraint and utilities, we get $u_{H} + u_{B} = 12$ . This is the efficiency frontier of the feasible set $S$ .
Bargaining Problem: Maximize the Nash Product $u_{H} \cdot u_{B}$ subject to $u_{H} + u_{B} = 12$ .

Since the problem is perfectly symmetric (as seen in the video's first example), the Symmetry axiom tells us the solution must be $u_{H}^{*} = u_{B}^{*}$ . This immediately gives us the solution:

$u_{H}^{*} = 6$ , $u_{B}^{*} = 6$ .
This corresponds to an equal time split: $t_{H} = 6$ hours, $t_{B} = 6$ hours.

This outcome is intuitive: when both players have identical preferences and bargaining positions, they split the surplus equally.

Case 2: Asymmetric Risk Preferences

Now, let's introduce a more nuanced and realistic assumption. Suppose the hotwife exhibits diminishing marginal utility for time; the first few hours are highly valuable, but each additional hour provides less satisfaction. This makes her risk-averse in the negotiation—she is more anxious to secure at least some agreement than to hold out for the absolute best deal. The bull, however, remains risk-neutral.

Constraint: $t_{H} + t_{B} = 12$
Utility Functions: $u_{H} = t_{H}$ (risk-averse) and $u_{B} = t_{B}$ (risk-neutral).
Feasible Set Boundary: We need to express the relationship between $u_{H}$ and $u_{B}$ .
- From the utility functions, $t_{H} = u_{H}^{2}$ and $t_{B} = u_{B}$ .
- Substituting into the constraint: $u_{H}^{2} + u_{B} = 12$ .
Bargaining Problem: Maximize $u_{H} \cdot u_{B}$ subject to $u_{H}^{2} + u_{B} = 12$ .

Following the calculus approach from the video, we substitute the constraint into the Nash Product:
Maximize $f (u_{H}) = u_{H} \cdot (12 - u_{H}^{2}) = 12 u_{H} - u_{H}^{3}$ .

To find the maximum, we take the derivative with respect to $u_{H}$ and set it to zero:

\frac{df}{d u _{H}} = 12 - 3 u_{H}^{2} = 0

3 u_{H}^{2} = 12 ⟹ u_{H}^{2} = 4 ⟹ u_{H}^{*} = 2

Now we find the corresponding values for the other variables:

Bull's Utility: $u_{B}^{*} = 12 - (u_{H}^{*})^{2} = 12 - 4 = 8$ .
Hotwife's Time: $t_{H} = (u_{H}^{*})^{2} = 4$ hours.
Bull's Time: $t_{B} = u_{B}^{*} = 8$ hours.

Interpretation:
The Nash Bargaining Solution allocates 4 hours to the hotwife's priorities and 8 hours to the bull's. The risk-averse player (the hotwife) receives a smaller share of the resource. As explained in the video, risk aversion can be interpreted as being less bold in negotiations. A risk-averse player is more willing to accept a smaller but certain gain, while a risk-neutral player can push harder for a larger share, less deterred by the risk of the negotiation failing. This demonstrates how the mathematical structure of utility functions directly impacts the predicted outcome of a fair negotiation.

5. The Power of Disagreement: Threats and Bargaining Power

Our model can be made even more sophisticated by adjusting two key elements:

Bargaining Power: What if the players are not symmetric? The Generalized Nash Bargaining Solution (GNBS) introduces exponents to represent relative bargaining power. The problem becomes maximizing $(u_{H} - d_{H})^{α} (u_{B} - d_{B})^{1 - α}$ , where $α$ is the hotwife's bargaining power (0 < $α$ < 1). If $α > 0.5$ , she has more influence, perhaps because she has better alternatives or more control over scheduling.
The Disagreement Point: We set $d = (0, 0)$ , but what if it's different? The theory of Rational Threats suggests that players might strategically choose actions that worsen the disagreement outcome for their opponent, thereby improving their own bargaining position. For example, if the bull could credibly commit to a strategy that would make disagreement unpleasant for the hotwife (while being only mildly inconvenient for him), this would shift the disagreement point $d$ in his favor and, consequently, shift the final Nash Bargaining Solution as well. This links the cooperative solution back to the non-cooperative, strategic thinking we've discussed previously.

Conclusion

In this lesson, we moved from dividing a pre-agreed pie to modeling how the agreement itself is reached. The Nash Bargaining Solution provides a powerful, axiomatically-grounded framework for predicting the outcome of a two-person negotiation.

Key Takeaways:

A bargaining problem is defined by a feasible utility set (S) and a disagreement point (d).
The Nash Bargaining Solution is the unique outcome that satisfies five axioms of fairness: Efficiency, Individual Rationality, Scale Covariance, IIA, and Symmetry.
This solution is found by maximizing the Nash Product, $(u_{1} - d_{1}) (u_{2} - d_{2})$ , which represents the product of the players' gains from cooperation.
Applying this to the hotwife-bull dynamic reveals that the final agreement is highly sensitive to the players' attitudes toward risk, as captured by their utility functions. A more risk-averse player will tend to receive a smaller share of the negotiated resources.

Having established frameworks for dividing gains (Shapley) and reaching agreements (Nash), we will now shift our focus to more operational challenges. The next module addresses scheduling and sequencing. We will begin by analyzing how to efficiently order a series of tasks, such as outfit changes in a photoshoot, to minimize total completion time.

Recommended resources