APPENDIXES - The Moral Economy: Why Good Incentives Are No Substitute for Good Citizens - Samuel Bowles

The Moral Economy: Why Good Incentives Are No Substitute for Good Citizens - Samuel Bowles (2016)



A Taxonomy of Additive Separability and Its Violations


Note: ΔT, ΔD, and ΔI are, respectively, the total, the direct, and the indirect effects of the incentive on the action, and ΔT = ΔD + ΔI.


Experimental Games Measuring Social Preferences and the Effects of Incentives




Source for the table: Adapted from Camerer and Fehr (2004)

Source for the games: Prisoner’s Dilemma: Dawes 1980 (survey); Public Goods: Ledyard 1995 (survey); Ultimatum: Güth, Schmittberger, and Schwarze 1982 (introduced the game), Camerer 2003 (survey); Dictator: Kahneman, Knetsch, and Thaler 1986 (introduced the game), Camerer 2003 (survey); Trust: Berg, Dickhaut, and McCabe 1995 (introduced the game), Camerer 2003 (survey); Gift Exchange: Fehr and Fishbacher 2001 (introduced the game); Third-Party Punishment: Fehr, Kirchsteiger, and Riedl 1993 (introduced the game).


Total, Direct, and Indirect Effects of the Subsidy in the Irlenbusch and Ruchala (2008) Experiment



Trust and the Liberal Rule of Law

This is a model of the relationship between the liberal state and trust, illustrating part of the argument at the end of chapter V.

Consider a population composed of a large number of people who interact in randomly selected pairs to engage in an exchange in which they may either behave opportunistically (for example, steal each other’s goods) or exchange goods to their mutual benefit. Call these strategies “defect” and “cooperate,” with payoffs describing a coordination game (also called an assurance game), as in the top payoff matrix in figure A.4. The structure of the game is such that if a player knows that the other will cooperate, then the payoff-maximizing strategy is also to cooperate (both then receive 4). But if the other is known to be a defector, then payoffs are maximized by also defecting (both then receive 2). The two equilibria are thus mutual defect and mutual cooperate (shaded cells in the payoff table, boxed payoffs in the right panel).

Expected payoffs for cooperators and defectors, which depend on the player’s subjective probability (p) that the other will cooperate, are labeled πC and πD in the right-hand panel. They are both increasing in (p). A player wishing to maximize her expected payoff will cooperate if she believes that the other will cooperate with at least some probability p*. This so-called critical value is determined by the intersection of the two payoff functions (to the right of p*, expected payoffs are higher if one cooperates). Because in the absence of the rule of law (thick lines in the figure), the critical value, p*, exceeds one-half, defection is termed the risk-dominant strategy, that is, it maximizes the expected payoffs of an individual who believes that his or her partner is equally likely to cooperate or defect. The rule of law (thin lines) reduces the gains of defecting on a cooperator and also reduces the cost to a cooperator if her partner defects. This lowers the critical value from p* to p. By requiring a smaller probability that the other will cooperate in order to motivate the player to cooperate, these changes make cooperation the risk-dominant strategy, and thus make cooperating easier to sustain.


Figure A.4. The rule of law and cooperative norms Left panel: Payoffs in the exchange game (upper without the rule of law, lower with the rule of law); the payoffs of the row player are the first entry in each cell. Right panel: Expected payoffs based on the type of one’s partner (heavy lines without the rule of law, thin lines with the rule of law).