An Experimental Research On Substitution Effect And Income Effect In The Choice Of Risky Assets

Substitution effect and income effect in Slutsky Equation reveal the essence of the interaction between price and preference. As it is impossible to observe substitution effect and income effect in the real life, the empirical test of Slutsky Equation and its amended form lags behind the theoretical research. Existing relative experimental researches rest on the test of indifference curve with oridinary goods as choosen objects, and haven't reached a uniform testing method. According to the revealed preference theory, we could get aware of subjects'preferences, even if their indifference curves are unknown. It is impossible to reveal subjects'real preference with ordinary goods as choosen objects, while possible with risky assets. Former consumer theories only focuses on the condition that consumers have complete ownership of the goods they buy, while ignores the principal-agent relationship which exists at large. As risky asset investment becomes more and more popular, studying substitution effect and income effect in both complete ownership frame and principal-agent frame with risky assets as choosen objects could not only test Slutsky Equation in the real life, but also provide theoretical support for investors'risky asset choices and the financial policy making of governance.This dissertation with risky assets as choosen objects, gets across the difficulties that ordinary goods couldn't reveal subjects'real preferences and indifference curve is hard to test, designs two risky assets with different risk levels, with the method of value inducing separates substitution effect and income effect by separately changing the prices of risky assets in the laboratory, and tests the relationship of the sum of substitution effect and income effect and price effect and their signs in the complete ownership and principal-agent frame respectively.The experimental results show that:(i) In the complete ownership frame, Slutsky Equation and its amended form are tenable in the choice of risky assets, while in the principal-agent frame, Slutsky Equation is tenable only if the price of risky asset with relative low risk level increases or that of risky asset with relative high risk level changes, while the amended Slutsky Equation is right only if the price of risky asset with relative low risk level decreases. (ii) The signs of substitution effect, income effect and price effect in complete ownership frame are in accordance with that in principal-agent frame. And substitution effect is non-positive, income effect is non-negative, and price effect is non-positive, which indicates that risky asset has the same preference character with ordinary goods. (iii) Subjects have the same risk preference in complete ownership frame and principal-agent frame. These results indicate that sufficient incentive mechanism does decrease the moral hazard in the principal-agent relationship, but couldn't eliminate it essentially, which proves that ownership ascription is the basic reason of moral hazard.There are three innovations in this research. Firstly, this research separates substitution effect and income effect from price effect for the first time. Secondly, by using the value inducing method, and taking risky assets as chosen objects, this research reveals subjects'real preference, which avoids the difficulties that indifference vurve is hard to test and ordinary good couldn't reveal subjects'real preferences. Lastly, this research studies Slutsky Equation in the principal-agent frame, and tests and compares the substitution effect and income effect in the complete ownership frame and principal-agent frame, so as to validate that the ownership ascription is the basic reason of moral hazard.