|
||||||||||
|
Table: Empirical
studies on ownership structure and performance[1]
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z |
||||||||||
|
Author(s) &Journal |
Sample
& Period[2]
|
Ownership |
Performance
variable(s) |
Other
variable(s): Controls & dependents[5] |
Statistical
methods |
Main
results |
Preferred
explanation |
|||
|
59,148 trading events by 769 public 1975-81. |
1) Insider sales by officers, directors and above 10% blockholders.
2) Insider purchases by officers, directors and above 10% blockholders. 3)
Outsider trading on public but delayed information on insider trades from the
Security Exchange Committee. 4) Outsider trading on public but delayed
information on insider trades from SEC's Official
Summery. |
1) Cumulative abnormal return CAR (adjusted for size effects) over
the intervals (TD0, TD+100) and (TD-100, TD0) where TD is the insider trading
date. 2) Cumulative abnormal return CAR further adjusted for transaction
costs and the bid-ask spread. |
1) Dummy for insider being a director. 2) Dummy for insiders that are
both officers and directors. 3) Dummy for insider being the chairman of the
board. 4) Dummy for insider being a 10% blockholder. 5) Dummy for insider
being an officer. 6) Log of $ value of trade. 7) Log of proportion of firm traded.
8) Log of value of firm. |
Event-study. GLS regression. For sales CAR is multiplied by -1 in
order for CAR to represent insider gains on trading. |
CAR increases significantly after insider purchases and it decreases
significantly after insider sales. CAR decreases significantly before insider
purchases and it increases significantly before insider sales. CAR increases
significantly after outsider trading at both announcement dates, but vanish
when transaction costs are added. GLS
regressions: CAR increases significantly for: 1) Insiders that are both
officers and directors. 2) Insider being the chairman. 3) Log value of trade.
4) % of firm traded. CAR decreases significantly for log of firm value. |
The insider-investment argument. |
||||
|
225 1988-92. |
1) % of shares held by directors. 2) % of shares held by institutions
with more than 5% ownership. 3) % external ownership. Source: Annual company reports. |
1) Return on equity. 2) Tobin’s Q by market value to book value. Source: Datastream. |
1) Size by sales. 2) Growth by sales growth. 3) Leverage by debt to
total assets. 4) R&D / total assets. |
Heteroskedasticity corrected OLS regression. Performance is regressed
as a 3rd degree polynomial of director ownership. |
Director ownership and cubic ownership is significantly positive and
squared ownership is significantly negative. The polynomial reach its maximum
at 16% and its minimum at 42% ownership. Significant controls: Size, growth,
and R&D. |
Morck et al.’s [1988] combined incentive alignment and entrenchment argument.
|
||||
|
Sample 1: 85 deaths of an officer or a member of the board of directors
owing at least 5% of the stock. Sample 2: 103 deaths of CEOs holding less than 5% of the stock. Both samples consists of publicly listed 1973-89 both samples. |
1) % shareholding by the deceased. 2) % shareholdings by non-deceased
officers and directors. 3) % ownership by insiders [officers and directors] before
and after the event. 4) % shareholdings by institutional investors. 5) Dummy
for deceased being a founder. 6) Dummy for deceased being a CEO. 7) Age of the
deceased. 8) Dummy for decline in % ownership by officers and directors. 9)
Dummy for no decline in % ownership by insiders and deceased shares pass to
multiple heirs. |
Two-day cumulative abnormal return CAR, by the firm over the interval
(AD-1, AD0) where AD is the announcement date. |
1) Dummy for corporate control bid. 2) Dummy for hostile control bid.
|
Event-study. Control for variables by sample classification. WLS
regression in order to correct for heteroscedasticity. |
Sample 1: 60% of the events leads to decreased ownership by insiders
and 52% receive bids for control within 10 years of the event. Using classified samples CAR increases significantly if the shareholding
of the deceased is above 10%. The rise in CAR is larger if ownership by the deceased
is between 20 and 30%, if the deceased is the founder, a non-CEO, or if ownership
by insiders decreases. WLS regressions show a significant bell-shaped relation between CAR
and the deceased shareholdings. However, none of the other ownership or control
variables are significant although they have the expected sign. Sample 2: No significant result. |
Stulz’s [1988] combined takeover premium and entrenchment argument. |
||||
|
Song and Walkling [1993], Journal of Financial and Quantitative Analysis |
459 |
1) % ownership by all officers and directors. 2) Interaction effect
for contested and managerial ownership. 3) Interaction effect of successful
and managerial ownership. 4) Interaction effect for contested, successful,
and managerial ownership. 5) Institutional ownership. Data from value line. |
Cumulative abnormal return, CAR, by the target firm over the interval
(A-5, R+5) where A is the announcement date of the acquisition activity and R
is the announcement date of the final outcome. |
OLS regression controls: 1) Dummy for contested. 2) Dummy for successfully
acquired. 3) Interaction effect for successful and contested. Logistic regression controls: 1) Size. 2) Leverage. 3) Liquidity. 4)
Valuation ratio. 5) Return on equity. 6) Price / Earnings. 7) Growth. |
Event study. Initially standard t-tests are applied to test for differences
in the three samples. Logistic regression with probability of acquisition attempt
as the dependent variable. OLS regression with CAR as dependent variable. |
CAR increases significantly with managerial ownership in contested
cases that are ultimately successful. Significant controls: Dummy for successfully
acquired. Other results: The probability of acquisition attempts is
significantly decreasing with managerial ownership at the target firm, and
the result is due to variance among contested targets. Finally, managerial
ownership is significantly lower in contested compared to uncontested acquisitions
and in unsuccessful compared to successful cases. |
The integrated takeover premium and entrenchment argument by Stulz
[1988]. |
|||
|
60 1948-66. |
MC<5% single block of ownership. OC³20%. |
1) Return on equity. 2) Stock return assuming dividends are reinvested.
3) Dividends / earnings. 4) Growth in sales. 5) Growth in equity. |
Classify the sample in industries by 11 three-digit SIC. |
Variance analysis. |
No average difference between MC and OC firms, but a few significant
differences between industry groups. |
The incentive argument. |
||||
|
354 of the largest 1963-72. |
MC £10% single block
of common stock. 30%>WOC >10%. SOC ³30%. AOC=WOC+SOC |
Stock return assuming dividends are reinvested, observed 1965-72. |
1) Growth of earnings per share. 2) Growth of sales. 3) Growth attributable
to mergers & acquisitions. 4) Beta risk. 5) Equity/assets. 6) Size by
assets. 7) Monopoly power. 8) Industry. |
Multiple regression analysis. Check for model specification error due
to dependence between sales growth and ownership type. |
SOC firms are significantly more profitable than MC firms. WOC are
not. The final regression that this conclusion is based upon does only control
for industry type. |
The incentive alignment argument. |
||||
|
82 large 15 in food, 8 in brewing, 12 in electrical, 23 in mechanical
engineering and 19 in distributive trades. 1967-71. |
OC ³15% of cohesive
ownership or ³3% ownership by
managers. MC otherwise. |
1) Return on equity. 2) Return on assets. 3) Return on sales. |
1) Organizational form. 2) Organizational change. 3) # Executives / #
board members. 4) Size by assets. 5) Growth. 6) Industry. 7) Sales / equity.
8) Debt / equity. |
OLS regression. Test for multicollinearity. |
OC firms are significantly more profitable (weak) than MC firms.
Organizational form and change are significant (the main interest of Steer
and Cable). Growth is significant. Other variables are insignificant. |
The incentive alignment argument. |
||||
|
Strickland, Wiles, and
Zenner [1996], Journal of Financial Economics |
|
|
|
|
Event study on the effect of monitor actions by small shareholders. |
Evidence that united small shareholder activism enhances shareholder value. |
|
|||
|
104 successful tender offers in the 1968-1986. |
Target ownership: 1) % managerial ownership by officers, directors
and other insiders prior to announcement. 2) % ownership by institutional
investors prior to announcement. 3) % ownership by bidder prior to announcement.
4) Bidder acquiring %. Data from Value Line, S&P's Stock Guide and |
Cumulative abnormal return, CAR by the target firm over the interval
(A-5, R+5) where A is the announcement date of the acquisition activity and R
is the announcement date of the final outcome. |
1) Market value of target equity. 2) Market value of bidder equity.
3) CAR if negative. 4) CAR if positive. |
Event study. Initially standard t-tests are applied to test for significance
of CAR in the total sample and the two sub-samples both with regard to bidder
CAR, target CAR and combined firm CAR. Then OLS regression with CAR as dependent
variable. |
Target CAR increases significantly with target managerial ownership
in successful, multiple bids. Target CAR decreases significantly with institutional
ownership in multiple bids and in total sample. Target CAR decreases significantly
with bidder ownership in total sample. Significant controls: CAR if positive in all samples; market value of
target equity for multiple bids. |
The takeover premium argument of Stulz's [1988] model. |
||||
|
Thomsen and Pedersen [1996], Management International Review |
600 firms selected among the 100 largest non-financial companies in 6
European countries. 1990-1993. |
Ownership concentration by logistic transformation of % voting
ownership by largest owner. |
Return on equity. |
1) Size by total assets. 2) Standard deviation of return on equity.
3) Dummy for public utilities. 4) Dummy for media companies. 5) Dummies for nation
effects. |
OLS regression. |
|
The natural selection’ argument. |
|||
|
Thomsen and Pedersen [1999], Journal of the Economics of Business |
518 firms selected among the 100 largest non-financial companies in
12 EU countries. 1990-1993. |
Ownership concentration by logistic transformation of % voting
ownership by largest owner. |
Return on equity. |
1) Size by total assets. 2) Standard deviation of return on equity.
3) Dummy for public utilities. 4) Dummy for media companies. 5) Dummies for nation
effects. 6) Dummy for dual class shares. 7) Size of economy by GDP. 8) Capital
intensity by assets to sales. 9) Dummy for R&D above 1% of corporate
turnover. 10) Stock market capitalization. 11) Bank sector concentration. |
OLS regression. |
Return on equity is insignificantly decreasing with ownership
concentration. Significant controls: Capital intensity and return
variability. Ownership concentration is not regressed against performance. Significant
controls: Firm size, standard deviation of return on equity, nation effect,
GDP, stock market capitalization and bank sector concentration. |
The natural selection’ argument. |
|||
|
Thonet and Poensgen [1979], The Journal of Industrial Economics |
Not altogether clear but probably this: 62 of 297 large German industrial firms. 1961-70. |
MC £25% of cohesive
stock ownership. OC ³25% of cohesive
stock ownership and no other part with 25% of cohesive stock ownership. |
1) Return on equity. 2) Stock return assuming dividends are reinvested.
3) Market value to book value. |
1) Firm size by assets. 2) Industry concentration ratios by market shares. Other dependents: 1) Sales growth. 2) Variance of return on equity. 3) Beta risk. |
Multiple regression: OLS, and GLS. Corrects for heteroscedasticity
due to size. |
OC firms are significantly less profitable than MC firms in terms of
return on equity and market value to book value. Size and concentration are
not significant. No significant difference between MC and OC with regard to
return on stocks, growth, variance on equity and beta risk. |
The incentive alignment argument. Thonet and Poensgen do not try to
explain why they can not confirm the incentive argument. |
|||
|
- Copyright 1997-2008, ViamInvest. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Legal notice. |
||||||||||
