way, and even fewer work like this all of the time. Behavioral economics is a collection of economics research that attempts to tackle this problem.
Herbert A. Simon introduced the concept of bounded rationality. It recognizes that decision makers’ information is incomplete. They might not fully understand all aspects of the decision itself. They might have limited time in which to make a choice, and not everyone’s mind works in the same way. Simon postulated that most decision makers use heuristic approaches; simple “mental shortcuts” instead of optimization.
Bounded rationality in general assumes that decision makers use techniques that lead to “good enough” results and stop optimizing further. This inherently takes into consideration the time factor of decision making.
These and other findings suggest that behavior is vastly more complex and harder to model than one expects, and that it depends on one’s current state of mind, psychological condition, and also one’s reference culture and other factors. One such aspect of human behavior is loss aversion—how individuals try to avoid losses more than obtain gains. This can be exploited to create compliance with contracts as explained in Chapter 6.
Handling Uncertainty
There is a large number of situations where the seller or the buyer has an incentive to hide certain information from the other, causing asymmetric information. In contracts for the sale of goods, there are some statutory provisions to prevent this from happening as discussed in Chapter 5. However, many situations are left unregulated for the parties to resolve. In this setting, some statistical analysis can be useful.
Expected Value
The first tool to resolve an asymmetric information problem is expected value. This simply involves summing the products of payouts with their expected likelihoods.
The general formula is the following: If the V event can take the values v1 … vN with the respective probabilities of p1 … pN, then the expected value of V is:
E(V)=∑i=0nvi*pi
Using this formula, the expected value of any event whose probabilities one knows (or can guess) can be determined. This, however, questions the point of insurance. The business success of insurance companies relies on their being able to predict with great accuracy the chances of a certain event happening. With the purchase of insurance comes the certainty that the insurance fee is higher than the expected value of the payout—otherwise the insurance company would go out of business, yet one purchases insurance regardless and considers it the prudent thing to do. Why?
Case Problem
Let’s assume you get offered the following game: Flipping a fair coin will earn you $5 if it lands on heads but nothing if it lands on tails. Playing the game once costs $2. Should you play this game?
Expected Utility
To understand the above dilemma, remember that consumers maximize utility and not gains. Most goods, money included, have diminishing marginal utility; an additional unit is less valuable the more there is of it. Together, these two insights explain the insurance phenomenon well. Relatively wealthy people (whose house hasn’t burned down, car has not been stolen, etc.) will value the same dollar amount less than the (post-tragedy) poor people. To put it in another way, even though one overpays for insurance compared to the likelihood of a damaging event, the event itself can be so disastrous, and the damage so daunting, that it provokes complete willingness to do so. Behavior like this is called risk averse behavior, and entities acting like this are called risk averse.
The opposite example is playing the lottery. The chance of winning is miniscule, the loss is fixed, and the expected value of the gains is usually lower than the cost. At the same time, the cost is usually bearable, and the mere prospect of winning is generally uplifting and mood-altering. The benefit here is twofold. The “rational” part is the expected value of the winning, while the other benefit is the excitement of gambling.
Interrelated Decision Making
One of the many advantages of the market model is that it allows us to ignore all other participants, as they are represented as an aggregate in the supply and demand curves. This makes it possible to focus on one’s own individual costs and benefits and then decide accordingly. In the real world, this model is best applicable in scenarios where the number of participants is large. There are, however, many situations where transactions happen in a one-on-one setting; in that case, the market model could be less helpful. Instead, one can use game theory, a framework developed specifically to analyze decisions in a strategic context. The benefits and costs do not depend just on one’s own actions but also on the actions of others.
Game Theory Basics
In game theory, a game consists of:
Players: The entities who take actions
Actions: All of the possible actions the players can undertake
Payoffs: The outcome of each possible combination of actions
Information: The information available to all players
Games can be cooperative or non-cooperative. A game is cooperative if there is an enforcement mechanism that ensures cooperative behavior. For the rest of the chapter, the focus is on non-cooperative games, to see how the legal environment can impact the outcome.
Simultaneous Games
In simultaneous games, the players have to choose an action before they are aware of the other players’ choices.17 These games (also called normal-form or strategic-form games) are commonly represented with a payoff matrix.18 One player is represented by the rows, the other by the columns, and each row/column represents a possible strategy. The cells of the matrix contain the payoffs for both players. (The first number represents the player in the rows, the second number the player in the columns.)
| Player B | |||
| Strategy B1 | Strategy B2 | ||
| Player A | Strategy A1 | 10, 2 | 9, 10 |
| Strategy A2 | 12, 3 | 11, 0 |
For example, in the game depicted above, there are two players, Player A and Player B. Both have two strategies, denoted A1, A2 for A, and B1, B2 for B. The inside of the matrix contains the payoffs; for example, if Player A plays A1 and Player B plays B1, the payoffs are 10 and 2, so Player A gets 10, and Player B gets 2.
If there is perfect information, both players are aware of all of the payout matrix possibilities and strategies available to each player. In a situation like this, the game is solved by analyzing each player’s potential actions.
Player A’s payouts depend on two things, his own choices and the choices of Player B. While Player A does not know what Player B wants, he can determine what the best choices for himself would be if Player B chooses B1 or B2. If Player B plays B1, then Player A could choose A1 for 10 but selecting A2 would yield him 12, which is better:
