Maximizing Impact_ Nonprofit Website Planning, Budgeting, and Design.pdf
Synopsis 1 h6
1. 1
COURSE TITLE: SEMINOR IN FINANCE COURSE CODE: MPH 622
Synopsis of article: The Cross-Section of Realized Stock Returns: The Pre-COMPUSTAT Evidence
By James L. Davas, Kansas State University, USA
Article published in: Journal of Finance, Vol. XLIXI No. 5 (December 1994), pp. 1579-1593
Submitted to: Prof. Dr. Radhe Shyam Pradhan, Masters of Philosophy in Management, TU
Submitted by: Sudarshan Kadariya, Roll No 04/’010, M. Phil II Semester
Issue of the study: Is there ability of certain variables to explain the cross-sectional variation in realized stock
returns?
Purpose: To find the explanatory power of book-to-market equity, earning yield, sales growth, firm size, stock price
and historical beta with respect to the cross-section of realized stock returns during the period from July 1940 to June
1963 and to find the January seasonal explanatory power of these variables.
Review of Literature: Relationship between book-to-market equity and stock returns by Rosenberg, Reid and
Lanstein (1985), De Bondt and Thaler (1987), Chan, Hamao and Lakonishok (1991) and Fama and French (1992),
earning yield and stock returns by Basu (1977), Jaffe, Keim and Waterfield (1989), cash flow yield and stock returns
by Chan, Hamao and Lakonishok (1991), historical sales growth and stock returns by Lakonishok, Shleifer, and
Vishny (1993), Seasonality is the explanatory power of earnings yield by Jaffe, Keim and Waterfield (1989) and
January seasonal in the magnitude of the regression coefficient on book-to-market equity by Fama and French
(1992).
Hypotheses: Based on the existing literature, author has developed three hypotheses which are based on the study of
Fama and French (1993), Lakonishok, Shleifer, and Vishny (1993) and Kothari, Shanken, and Sloan (1993)
respectively. H1: Certain fundamental variables are measuring the riskiness of stocks, so that the correlation between
the variables and subsequent returns reflects compensation for bearing risk. H2: The variables allow investors to
identify stocks that are mispriced, thus creating opportunities for realized returns in excess of what is required to
compensate investors for risk and H3: The observed predictive ability is an artifact of the research design and
database used to conduct the study, and the predictive ability of certain variables would be reduced or vanish if
different methodology and data were used. This study addresses some of the methodological issues that have been
raised in connection with this H3 hypothesis and H1 & H2 explain the predictive ability of the variables.
Limitations: i) This study is based on the largest firms of NYSE and extremely low-priced stocks are virtually
excluded from the sample. Since, the study is focuses on the firms in the top half of the size spectrum so that
problems with infrequent trading and bid-ask spreads are less likely to affect the results. ii) The ratio of book-to-
market equity was calculated on the basis of accounting information that appears in a firm’s financial statements
rather than market value of equity. iii) Calculations of selected variables were based on the accounting information.
Therefore, the result of the study may be affected by the COMPUSTAT related problems. But, the survivorship bias
in the selection of firms, Davis (1994), Banz and Breen (1986) and problem of a look-ahead bias, Banz and Breen
(1986) out of four problems related to the use of COMPUSTAT data have been eliminated due using Moody’s
Industrial Manuals as data source. Moreover, Data snooping, Lo and MacKinlay (1990) and short sample period (the
COMPUSTAT is limited to a relatively short sample period because comprehensive data are available only after
1963) still remains in this study.
Data sources: This study was based on the primary sources of data taken from Moody Industrial Manuals - for book
value earnings, cash flow, and sales and University of Chicago, Center for Research in Security Prices (CRSP) - for
stock returns, stock prices and market values of equity.
2. 2
Sample and study period: 100 firms as sample were selected on the basis of top half of the June 30 size ranking and
listed in the Moody’s Industrial Manual. Reasons for preferring starting date are; first, the period from 1940 to the
early 1960s is a period of fairly constant volatility in the stock market compare with 1930s and availability of the
accounting information.
Variables: Book-to-market equity, cash flow yield, earning yield and historical sales growth are of primary focus of
the study rather other variables; BETA: coefficient assigned to each stock according to their historical betas, firm
size, share price, LBM: natural log of the ratio of book to market value of equity, LMV: natural log of market value
of equity, E/P: ratio of earning per share to stock price, CF/P: ratio of cash flow per share to stock price, GROWTH:
five year compound annual average sales growth rate and LPRICE: natural log of the stock price as of each June 30
for each firm also included in the analysis.
Procedures of analysis: The empirical analysis takes two forms. First, the stocks in the sample were ranked by the
variables and quintiles were formed based on these rankings. Second, the returns, systematic risk, and other
characteristics of these quintiles were analyzed. In addition to rankings, portfolios were also formed through two
ways. First portfolio was formed that consists of stocks that are in the top third of the cash flow yield ranking and in
the bottom third of the sales growth ranking and the second portfolio was formed that consists of stocks that are in
the bottom third of the cash flow yield ranking and in the top third of the sales growth ranking. The overall
performance of the portfolios is considered as the worst and best, respectively as per Lakonishok, Shlefer, and
Vishny (1993). The provision of the return on the delisted stock during the year (study period) was considered equal
to the average of the returns to the two stocks nearest to the delisted stock in the appropriate ranking. Another part of
the analysis involves a Fama-MacBeth (1973) cross-sectional regression analysis to determine which factors have
explanatory power with respect to the cross section of realized returns. Monthly regressions were run during the
study period from July 1940 to June 1963. The time series average and standard deviation of each series of
parameters were used to assess the statistical and economic significance of the independent variables. The results of
the monthly regressions were presented for all months, for January only, and for February-December only as per the
objectives of this study.
Empirical results: Portfolio results; i) Ranking on the basis certain variables produces dispersion in returns. For
instance, E/P and CF/P produce annual return differences more than 5% (specifically, more than 9.5% and 6.8%
respectively). None of the other ranking produce more than 5% differences, however 3.1% in firm size. (ref. table I),
ii) There are correlations among the variables. For example E/P and CF/P are highly correlated, CF/P and B/M value
of equity because there variables are the scaled version of price. (ref. table I) and comments regarding the negative
E/P and CF/P portfolios are due to 14 of the 23 years, no firms have negative earnings. Similarly, no firms have
negative CF in 17 of the years. In the remaining years, these portfolios typically contain few stocks with negative
earnings and CF. (ref. table I). Regression results; i) There is a statically significant relationship between book-to-
market equity and subsequent returns. (ref. table II), ii) Cash flow yield also has explanatory power respect to
subsequent realized returns when book-to-market equity and historical sales growth are held constant. (ref. table II),
iii) Earning yield has also explanatory power in regression analysis. iv) There is insignificant explanatory power for
beta. (ref. table II), v) There is a weak relationship between sales growth and return. (ref. table II) and Seasonal
patterns; i) There is significance of LBM, E/P+, and CF/P+ with return mostly in January. Therefore, January
seasonal is not restricted to small firms during the study period. (ref. table III)
Conclusions: Book-to-market equity, earning yield, and cash flow yield have significant explanatory power with
respect to the cross-section of realized stock returns during the study period and there is a strong January seasonal in
the explanatory power of book-to-market equity, earning yield and cash flow yield.
Critical appraisal: The study has found out the relationship between the variables which were already identified,
thus it is believed that the work was conducted for the conformation of the earlier studies and contains of the article
is quite repetitive. ***