IFSS_Rationalism_Sheng

IFSS (STRE 8005) Semester 1, 2009 Positive Economics summarized by Sixin Sheng (z3273424) Friedman, M. (1953) “The methodology of positive economics”, in Essays in Positive Economics, Chicago: University of Chicago Press, 3- 43. In this influential paper, Friedman firstly discussed the differences between positive economics and normative economics, and then answered at least four important questions: firstly, what is the essence of the theory? Secondly, could we refuse a theory based on its unrealistic assumptions? Thirdly, which roles do a theory’s assumptions play? Finally, how could we identify a better theory among alternatives? Different from normative economics, positive economics is “independent of any particular ethical position or normative judgements”. Positive economics is dealing with the question of “what is” and making predictions, while normative economics answers the question of “what ought to be” and makes policy recommendations. In a way, positive economics can provide normative economics grounds and direction for its policy conclusions. Friedman argued that positive economics is to develop a theory that can predict unobserved phenomena. On the one hand, theory can be viewed as a language that aims to “promote systematic and organized methods of reasoning”, but itself does not include any substantive content. In that sense, theory should be evaluated by the logic of its language (simplicity and completeness); On the other hand, theory also can be viewed as “a body of substantive hypothesis” with some certain explanatory and predictive powers. In this sense, only factual evidence can judge a theory’s correctness or validness. It should be noted that, when we test a theory, the logical criterion is relevant but its role is just subsidiary. Some people criticized that the assumptions of economic theories (such as maximization of returns assumption) are often not realistic, so they thought economic theories with unfounded assumptions should be rejected. However, Friedman believed that the assumptions of a theory are not necessary to be realistic (completely realistic assumptions are impossible as well), because the realism of the assumptions cannot play any role in testing a theory. The only relevant test of a theory’s validity is the “comparison of its predictions with factual evidence”. For Friedman, a theory can be falsified by contradicted factual evidence, but a theory cannot be proved by limited factual evidence consistent to its predictions, which just can improve our confidence on the theory’s validity. If so, why do we use or need assumptions in many theories? Friedman said the assumptions of a theory may still have three positive roles. Firstly, they can be “an economic model of describing or presenting a theory”. Secondly, they may facilitate “an indirect test of the hypothesis by its implications”. Thirdly, they also can be “a convenient means of specifying the conditions under which the theory is expected to be valid”. Finally, Friedman pointed out the complete realism of a theory is not possible, and a theory is better than its alternatives only if this theory’s predictions can satisfy our research purposes better, or this theory can yield more accurate and widespread predictions than other alternative theories. In order to promote positive economics, Friedman thought it is necessary to test and refine existing theories. In addition, constructing new theories continually is also very important. IFSS (STRE 8005) Semester 1, 2009 Bounded Rationality summarized by Sixin Sheng (z3273424) Simon, H. (1978) “Rationality as Process and as Product of Thought”, American Economic Review, 68(2), 1-16. Some economists defended their rational man assumption using a narrow definition of rationality: utility maximization in the allocation of scarce resources, while other social sciences such as sociology proposed a broader usage of rationality, that is, human behaviour is rational. However, through a detailed illustration of how rationality entering into other social sciences in the form of functional analysis, Simon in his lecture argued that there is no much substantial difference between the views of rationality in economics and its sister social sciences. Thus, inter-disciplinary study and communication within different social sciences should be possible and helpful for a better understanding of human behaviour. Simon agreed that, rational component is quite common in almost all human behaviour, if the term rational taking its broader everyday sense. Even for economics, this broader version of rationality is actually used very often in much economic literature, but for many social sciences, the problem is that the dominant form of reasoning in them relies on a quite simple causal analysis that pays much attention to qualitative and structural questions but does not pursue mathematic calculation of optimal balance. In the similar vein, Simon argued that economics needs to shift its focus from the results of rational choice to the process of choice. In that sense, economic account should attach more importance to procedural rationality, i.e., the effectiveness of the procedures used to make choice, based on human intellectual powers and limitations. As a result, rational behaviour theories are required to consider both the means the actors used to deal with “uncertainty and cognitive complexity” and “the characteristics of the objective environment” for their decisions. A fact is that our attention as scarce resource can only deal with limited information related to complex problems. Given this fact, a critical question is how people make choice in light of limited information and limited computational capacity, and that was what Simon emphasized the issue of “human bounded rationality” in the choice process. One common way to make rational choice is try to search all the alternatives, however, because of time and cognitive constraints, actors generally must choose an action before all possible alternatives have been found. Another way resorts to the team work in which decisions can gain more information from communications among team members, but the team members’ cognitive limitations and limited attention still remain an issue. In addition, people may use various advanced technologies/knowledge such as artificial intelligence to improve rational search procedures, yet quite a few problems’ complexity is beyond the reach of those technologies/knowledge. In a word, any proposed “rational solution” to the problem is incomplete and just an approximation to the real situation. In some cases, the solution even can be totally deviated from the reality. In conclusion, people’s capacity to achieve procedural rationality is advanced with the help of various inventions and equipment, although absolute rational choice is always impossible. In order to complete rational choice theory, Simon called for a multi-disciplinary research method to build a better theory about the procedural rationality.