本章目录 《博弈与社会》第1章 社会问题 导论:个人理性与社会效率 ·理性人行为 ·效率标准 张维迎教授 外部性与科斯定理。 北京大学光华管理学院 社会的两个基本问题 协调问题:以交通为例 ·社会是什么 靠左行靠右行 协调( coordination) 合作( cooperation) l,1 协调与预测 合作问题 ·协调问题的核心是预测( prediction) 不合作 预测其他人如何行为 ·个人行为的基本规则是什么? 3,3-1,4 ·信息: too little knowledge, too much knowledge fail to use the knowledge, no knowledge can help 4,-10,0
1 《博弈与社会》第1章 导论:个人理性与社会效率 张维迎 教授 北京大学光华管理学院 本章目录 • 社会问题; • 理性人行为; • 效率标准; • 外部性与科斯定理。 社会的两个基本问题 • 社会是什么? • 协调(coordination) • 合作(cooperation) 协调问题:以交通为例 靠左行靠右行 靠左行 靠右行 A B 1, 1 1, 1 -1, -1 -1, -1 协调与预测 • 协调问题的核心是预测(prediction): – 预测其他人如何行为; • 个人行为的基本规则是什么? • 信息:too little knowledge; too much knowledge; fail to use the knowledge; no knowledge can help. 合作问题 合作不合作 合作 不合作 A B 3, 3 0, 0 -1, 4 4, -1
合作与激励 协调与合作相互作用 ·合作问题的核心是激励( Incentive ·如交通问题 个人行为与社会利益的冲突 ·如密云元宵节惨剧; ·什么是个人的最优选择? ·如SARS问题,禽流感问题 ·什么是社会的最优选择? ·如恐怖主义问题 ·如何把个人行为与社会利益统一起来? 如股票市场 ·如宏观经济波动; ·等等。 社会问题及其解决办法 理解人的行为 ·社会规模:小社会,大社会,全球化 ·所有社会科学都是有关人的行为的科 技术问题 学,但不同学科的方法、假设、着眼点 社会分工 不同 ·社会制度:个人的行为规则 ·经济学:从个人到社会( from micro 正式制度;明规则,法律 非正式制度;潜规则,社会规范,文 社会学:从社会到个人 from macro to 价格制度; micro 非价格制度 ·心理学:用观察、实验的方法看人们实 际上是怎么行为的。(行为科学 Homo economicus vs Homo sociologicus Adam Smith ys. Emile durkheim The former adapts to changing circumstances, Rational. outcome-oriented: social norm always on the lookout for improvement. The latter oriented not outcome-oriented nsitive to circumstances. sticking to the escribed behavior even if new and apparently "the former is supposed to be guided by better options become available. The former is nstrumental rationality, while the behavior of the easily caricatured as self-contained, asocial atom latter is dictated by social norms. The former is and the latter as the mindless playing of soci pulled' by the prospect of future rewards forces. "(Jon Elster, 1989, JEP) quasl-inertial forces
2 合作与激励 • 合作问题的核心是激励(incentive) – 个人行为与社会利益的冲突 • 什么是个人的最优选择? • 什么是社会的最优选择? • 如何把个人行为与社会利益统一起来? 协调与合作相互作用 • 如交通问题; • 如密云元宵节惨剧; • 如SARS问题,禽流感问题; • 如恐怖主义问题; • 如股票市场; • 如宏观经济波动; • 等等。 社会问题及其解决办法 • 社会规模:小社会,大社会,全球化; • 技术问题: • 社会分工: • 社会制度:个人的行为规则 – 正式制度;明规则,法律 – 非正式制度;潜规则,社会规范,文化 – 价格制度; – 非价格制度。 理解人的行为 • 所有社会科学都是有关人的行为的科 学,但不同学科的方法、假设、着眼点 不同; • 经济学:从个人到社会(from micro to macro) • 社会学:从社会到个人(from macro to micro) • 心理学:用观察、实验的方法看人们实 际上是怎么行为的。(行为科学) Homo economicus vs. Homo sociologicus • Adam Smith vs. Emile Durkheim • Rational, outcome-oriented; social normoriented, not outcome-oriented: • “the former is supposed to be guided by instrumental rationality, while the behavior of the latter is dictated by social norms. The former is ‘pulled’ by the prospect of future rewards, whereas the latter is ‘pushed from behind by quasi-inertial forces. Cont’ • The former adapts to changing circumstances, always on the lookout for improvement. The latter is insensitive to circumstances, sticking to the prescribed behavior even if new and apparently better options become available. The former is easily caricatured as self-contained, asocial atom, and the latter as the mindless playing of social forces.” (Jon Elster, 1989, JEP)
为什么学习博弈论? Contd Nash Equilibrium has had a fundamental and ·博弈论是有关“互动行为”( Interactive pervasive impact in economics and the social behavior)的科学 science which is comparable to that of the A sort of umbrella, or unified field theory for the discovery of the dna double helix in the rational side of social science. It develops iological sciences. Nash formulated the basic methodologies that apply in principle to all ocabulary for a new language of economic nteractive situations. (Aumann and Hart, 1992) analysis." (R Myerson, 1999JEL) It provides solid micro-foundations for the study More and more, the language of game theory has of social structure and social change. (Jon Elster become the language of economics. More and 1982) more,'pure'theory came to mean game theory. (Faruk Gul, 1997, JEP) 博弈论的基本假设 关于理性人假设 ·理性假设( Rationality): agents are ·理性人的定义:有一个很好定义的偏好 ( well-defined preference在给定约束下 ·共同知识( common knowledge of 追求偏好的最大化( maximization) rationality) ·偏好是如何形成的 Know the rules of the game 关于偏好的假设 完备性( completeness):任何两个可选择的 方案之间都是可比较的 传递性( transitivity)(偏好具有一致性) 偏好函数 无差异曲线 ·如果偏好满足其他一些假设(特别是连 续性假设),可以定义一个偏好函数或 称效用函数( utility function) U=f(x, y) Lexicographic preferences · endowment effect
3 为什么学习博弈论? • 博弈论是有关“互动行为”(interactive behavior)的科学 • “A sort of umbrella, or unified field theory for the rational side of social science. It develops methodologies that apply in principle to all interactive situations.”(Aumann and Hart, 1992) • “It provides solid micro-foundations for the study of social structure and social change.”(Jon Elster, 1982) Cont’d • “Nash Equilibrium has had a fundamental and pervasive impact in economics and the social science which is comparable to that of the discovery of the DNA double helix in the biological sciences. Nash formulated the basic vocabulary for a new language of economic analysis.”(R. Myerson, 1999,JEL) • “More and more, the language of game theory has become the language of economics. More and more, ‘pure’ theory came to mean game theory.” (Faruk Gul, 1997, JEP) 博弈论的基本假设 • 理性假设(Rationality): agents are instrumentally rational; • 共同知识(common knowledge of rationality) • Know the rules of the game 关于理性人假设 • 理性人的定义:有一个很好定义的偏好 (well-defined preference);在给定约束下 追求偏好的最大化(maximization) • 偏好是如何形成的? • 关于偏好的假设: – 完备性(completeness):任何两个可选择的 方案之间都是可比较的; – 传递性(transitivity):(偏好具有一致性) 偏好函数 • 如果偏好满足其他一些假设(特别是连 续性假设),可以定义一个偏好函数或 称效用函数(utility function); • Lexicographic preferences • endowment effect. U = f (x, y) 无差异曲线 X Y A B
约束条件 确定环境下理性人行为的基本特征 ·技术性约束 最优选择由偏好和约束条件决定:人们 如一天工作不能超过24小时 选择的不同可能是由于偏好不同,也可 制度性约束: 能是由于面临的约束条件不同,如买 如晚上12点必须熄灯; 车 ·信息约束 ·最优选择在个人边际成本等于边际收益 如你并不知道这门课有什么价值 时达到 对他人行为的预期 ·最优选择会随外生变量的变化而改变。 不确定环境下的最优选择 不确定性与期望效用函数 预期效用理论( ·不确定性结果的效用可以表达为确定性 期望效用理论( prospect theory) 结果的期望值:效用函数对概率是线性 的 )=∑pU(x) 对理性人假设的批评 PROSPECT UTILITY THEORY ·有限理性( bounded rationality) intended to Reference-dependent utility:效用水平定义 be rational, but only limited to be so 在财富的变化上,而非财富的最终状态 上; and serious flawed memories ·有限毅力( bounded willpower) framing problem:认知心理学 Present-biased preference non-linear:风险(概率)是非线性的 有限自利( bounded self- Inter emotional behavior
4 约束条件 • 技术性约束: – 如一天工作不能超过24小时; • 制度性约束: – 如晚上12点必须熄灯; • 信息约束: – 如你并不知道这门课有什么价值; – 对他人行为的预期 确定环境下理性人行为的基本特征 • 最优选择由偏好和约束条件决定:人们 选择的不同可能是由于偏好不同,也可 能是由于面临的约束条件不同,如买 车; • 最优选择在个人边际成本等于边际收益 时达到; • 最优选择会随外生变量的变化而改变。 不确定环境下的最优选择 • 预期效用理论(expected utility theory) • 期望效用理论(prospect theory) 不确定性与期望效用函数 • 不确定性结果的效用可以表达为确定性 结果的期望值:效用函数对概率是线性 的: ( ) = ∑ ( ) i i V P pU x 对理性人假设的批评 • 有限理性(bounded rationality):intended to be rational, but only limited to be so. – Cognitive ability(limited computational skill and serious flawed memories) • 有限毅力(bounded willpower); – Present-biased preference • 有限自利(bounded self-interest) – emotional behavior PROSPECT UTILITY THEORY • Reference-dependent utility:效用水平定义 在财富的变化上,而非财富的最终状态 上; • framing problem:认知心理学; • non-linear:风险(概率)是非线性的
风险决策 效用曲线 ·输赢的不对称:当人们面临风险决策 时,他们更在乎的是成功与失败,赢和 输,财富的变化,而不是最终的财富状 态;比如说,100元损失导致的效用的减 少远远大于100收益导致的效用的增加 人们是损失规避型的( oss aversion) 效用是“禀赋(占有)依存”的 (endowment effect; status quo bias) 个例子 基本结论 ·第一组(76人):每人得到一个咖啡 ·人们最在乎的是他们已经得到的东西 杯;然后问他们是否愿意用咖啡杯换一 “借锅容易拔锅难”。 块400克的巧克力:89%的人不愿意 ·占有的时间越长,失去的痛苦越大。 ·第二组(87人):每人得到一块400克的 巧克力;然后问他们是否愿意用手中的 巧克力换一个咖啡杯:90%的人不愿 Framing problem 对理性人假设的捍卫 问题1(N=152)设想一种不常见的疾病预期将 Second best: no reliably accurate and analytically 导致600人的死亡。如果方案A被采纳,200人 tractable theory of inconsistence and foolishness 的生命将得救(72%):如果方案B被采纳 has been develo 有1/3的可能性600人都得救,2/3的可能性没有 Survival: peoples behavior is expected to more 人得救(28%):(参考点:600人死) closely approximate the ideal rationality ·问题2(N=155):如果方案C被采纳,400人 ormative: functional goal of social science is not 将会死亡(22%):如果方案D被采纳,有13 的概率400人会死亡,2/3的概率600人会死亡 just to predict human behavior, but also to analyze social institutions and evaluate proposals for (参考点:没有人死) institutional reform
5 风险决策 • 输赢的不对称:当人们面临风险决策 时,他们更在乎的是成功与失败,赢和 输,财富的变化,而不是最终的财富状 态;比如说,100元损失导致的效用的减 少远远大于100收益导致的效用的增加; • 人们是损失规避型的(loss aversion); • 效用是“禀赋(占有)依存”的 (endowment effect; status quo bias)。 效用曲线 损失 收益 效用 一个例子 • 第一组(76人):每人得到一个咖啡 杯;然后问他们是否愿意用咖啡杯换一 块400克的巧克力:89%的人不愿意; • 第二组(87人):每人得到一块400克的 巧克力;然后问他们是否愿意用手中的 巧克力换一个咖啡杯:90%的人不愿 意。 基本结论 • 人们最在乎的是他们已经得到的东西: • “借锅容易拔锅难”。 • 占有的时间越长,失去的痛苦越大。 Framing Problem • 问题1(N=152)设想一种不常见的疾病预期将 导致600人的死亡。如果方案A被采纳,200人 的生命将得救(72%);如果方案B被采纳, 有1/3的可能性600人都得救,2/3的可能性没有 人得救(28%);(参考点:600人死) • 问题2(N=155):如果方案C被采纳,400人 将会死亡(22%);如果方案D被采纳,有1/3 的概率400人会死亡,2/3的概率600人会死亡。 (参考点:没有人死) 对理性人假设的捍卫 • Second best: no reliably accurate and analytically tractable theory of inconsistence and foolishness has been developed; • Survival: people’s behavior is expected to more closely approximate the ideal rationality; • Normative: functional goal of social science is not just to predict human behavior, but also to analyze social institutions and evaluate proposals for institutional reform