The Study of Excess Returns of the Liquidity Risk of each

Sector in Chinese Stock Market 李莉莉，张玉兰LI Lili,ZHANG Yulan

I. Introduction

Foreign studies:

Liquidity risk can lead to excess returns on the stock market.

Chinese stock market:Researchers cannot agree with each other on excess returns coming from liquidity risk.

II. Empirical Analysis

A. Liquidity indicator:

1/h l ct t t

tt

P P PNL

T

II. Empirical Analysis

B. Data selection:Bull phase: July 1, 2006 - October 16,2007 (the Shanghai A-share Index rose to 6395.76 points from 1784.46 points)

Bear phase: October 16, 2007 - December 31, 2008 (the Shanghai A-share Index fell to 1911.79 points from 6395.76 points)

II. Empirical Analysis

C. Empirical process and results:

Panel data model:

1 1 2 2 3 3 itit it it it it it it itY X X X

II. Empirical Analysis

C. Empirical process and results:

Effects Test Statistic d.f. Prob.

Cross-section F 1.2741 -299170 0.1480

Cross-section Chi-square 37.0034 29 0.1461

Table 1.Redundant Fixed Effects Tests (C:BULL PHASE)

II. Empirical Analysis

C. Empirical process and results:

Table 2.Correlated Random Effects - Hausman Test (C:BULL PHASE)

Test Summary Chi-Sq. Statistic Chi-Sq. d.f. Prob.

Cross-section random 18.6924 2 0.0001

II. Empirical Analysis

C. Empirical process and results:

Table 3.Estimation of Model Coefficient (C:BULL PHASE)

Variable Coefficient Std. Error t-Statistic Prob.

C 0.0043 0.0008 5.49 0.00

-0.0806 0.0189 -4.26 0.00

-0.0062 0.0016 -3.84 0.00 13

II. Empirical Analysis

C. Empirical process and results:

We find :

1.Individual fixed effects

2.Liquidity premium does not exist.

3.Value effect does not exist.

II. Empirical Analysis

C. Empirical process and results:

Table 4.Estimation of Model Coefficient (C: large-cap, bull phase)

Variable Coefficient Std. Error t-Statistic Prob.

C 0.0032 0.0012 2.75 0.01

-0.0007 0.0210 -0.03 0.97

-0.0046 0.0025 -1.85 0.06

1

3

II. Empirical Analysis

C. Empirical process and results:

Table 5.Estimation of Model Coefficient (C: medium-cap, bull phase)

Variable Coefficient Std. Error t-Statistic Prob.

C 0.0077 0.0015 5.23 0.00

-0.2530 0.0485 -5.22 0.00

-0.0075 0.0027 -2.73 0.01

1

3

II. Empirical Analysis

C. Empirical process and results:

Table 6.Estimation of Model Coefficient (C: small-cap, bull phase)

Variable Coefficient Std. Error t-Statistic Prob.

C 0.0046 0.0016 2.95 0.00

-0.4132 0.0615 -6.72 0.00

-0.0008 0.0034 -0.23 0.82

1

3

II. Empirical Analysis

C. Empirical process and results:

Manufacturing sector (Bull phase):

There are no liquidity risk premium and value effects for the small-cap, medium-cap, large-cap stocks.

II. Empirical Analysis

C. Empirical process and results:

we similarly analyze other industry sectors, the large, medium and small cap both in the bull phase and the bear phase.

We find : The liquidity risk premium does not exist and there is a certain degree of negative correlati-on of the liquid risk and the excess return.

II. Empirical Analysis

A. Liquidity indicator:

1/h l ct t t

tt

P P PNL

T

Among them, refers to the highest stock price for the day, the lowest stock price, the closing stock price the day before, the turnover for the day.

htP

ltP

1ctP

T

III. Summary

More effective measures to manage liquidity risk:1. Further improve the trading mechanism ; 2. Strengthen the laws and regulations related to liquidity risk; 3. Nurture powerful institutional investors to improve risk control system; 4. Further build information disclosure system; 5. Develop the philosophy of rational investment and reduce the incidence of “herding”.

Thank you!