Description
Research Proposal On The Incentive Effect Of Employee Ownership: Another Measurement And Its Determinant:- An employee stock ownership plan (ESOP)[1] is an employee-owner scheme that provides a company's workforce with an ownership interest in the company. In an ESOP, companies provide their employees with stock ownership, often at no up-front cost to the employees.
Research Proposal On The Incentive Effect Of
Employee Ownership: Another Measurement And
Its Determinants
Abstract:
Noticing the limitations in most of relevant empirical studies, this research tries to
develop a new method to (1): test whether the corporate performance change led by the
adoption of ESOP per se is significant, and (2): figure out the core determinants for the
incentive effect of equity ownership. After appropriate selection of proxy variables for
signifying firms' inputs and outputs, DEA approach will be employed to evaluate firm's
efficiency change. Then the regression models will be built to achieve the above two
goals.
1
Contents
1.BACKGROUND.................................................................................................................................... 3
2.LITERATURE REVIEW....................................................................................................................... 3-4
3.MOTIVATION................................................................................................................................... 4-6
3.1To Compute the Performance of the firm concerning Workers
Efficiency................5
3.2 To Develop a Method Calculating the Part of Productivity Change Caused
By the Adoption of ESOP
Solely.....................................5-6
3.1To Discover the Crucial Determinants for an Effective
ESOP......................6
4.RESEARCH SIGNIFICANCE................................................................................................................6-7
5.METHODOLOGY............................................................................................................................ 7-12
5.1 Ingredient Variables Selection—Step
1....................................7 5.2 Measurement of
Performance: DEA Method—Step2........................7-8 5.3
Measurement of Performance Change Caused by ESOP—Step
3................8-9 5.4Empirical Test and the Factors Determining the
Effectiveness of ESOP—Step 4......10-12
6.DATA................................................................................................................................................. 12
7.REFERENCES................................................................................................................................13-14
2
1. BACKGROUND
Employee stock ownership plans (ESOPs) have received much attention in the popular
business press as well as the academic field of corporate finance (Borstadt and
Zwirlein(1995)). In a specific ESOP, non-executive employees are given chances to own
stock, which makes it possible for them to hold some residual claims on the company
like shareholders, thereby the so-called "people's capitalism" is expected to be achieved.
The supporters of ESOPs usually stress its incentive effect on workers, whose argument
is that through ESOP participation, employees will be more motivated and begin to
think and act as owners. They will exercise fiduciary-like responsibility over the
resources of the firm, minimize agency costs, and align their behavior with the goals of
the firm (Henry et al.(2007)). If this is true, it is reasonable to anticipate that an ESOP-
based firm would outperform comparable firms in its industry group. Such hypothesis
has been supported by many empirical works concerning the firms all around world. But
there are also opposite viewpoints against the incentive hypothesis of ESOPs. Using
different samples or statistical methods, some empirical evidences also cast doubts on
the validity of the proposition that the adoption of ESOPs can improve corporate
performance significantly.
Without consistent empirical results, how employee ownership actually affects workers'
productivity and corporate performance is still an open question. Whether and under
what conditions such practices make sense or are really ill-advised should be found by
appropriate approach. This research tries to develop a method to investigate this old
question with a new perspective.
2. LITERATURE REVIEW
The basic rationale for employee ownership is that it leads to the convergence of
interests employees' interests with those of shareholders, which inspire workers to work
hard for achieving better corporate performance (Vanek(1965), Drucker(1978) and
Aoki(1984)). However, empirical results about the incentive hypothesis of employee
ownership are ambiguous. A detailed studies review about employee ownership and
corporate performance can be found in The National Center for Employee Ownership
(NCEO)
1
. While some studies show positive relationship between them (such as
Kruse(1992), Kato and Jones(1995)and Kim(2010)), there are also studies render an
opposite conclusion(such as Faleye et al.(2006)). But general speaking, most of
1
"Research on Employee Ownership, Corporate Performance and Employee Compensation",
http://www.nceo.org/main/article.php/id/3/
3
empirical results provide mixed evidences (Weston et al.(2004)). For example, Park and
Song(1995) finds that though firms with outside block holders adopting ESOP
experience a permanent improvement in performance, ESOP makes the firms'
performance without block holders declines. Conte et al.(1996)finds that financial
returns of public companies with ESOP are higher than those of comparable non-ESOP
companies but decline after the adoption of ESOP. Borstadt and Zwirlein(1995)finds
some evidences that ESOPs established as an incremental employee benefits results in
productivity improvement. However, after adjustment for industry data, no significant
change of performance is observed. Pugha et al.(2000)finds that few of financial
performance do improve significantly after the adoption of ESOP, but this appears to be
largely short-term.
The inconsistency of empirical results also leads scholars to cast doubts on the validity
of incentive effect of employee ownership. Some other theories about the consequence
of employee ownership have been proposed. In the view of investment and portfolio
selection, Muelbroek(2002) argue that any type of employee ownership all violates the
basic principle of portfolio diversification, which would make workers worse-off. From
this sense, ESOPs will not form an effective stimulus to inspire workers but just expose
them to unnecessary financial risks. Beatty(1995), Pagano and Volpin(2003)constructs
another counterargument against employee ownership that the adoption of ESOP
usually serves as a determent to hostile takeover because workers are stable
shareholders who would not be likely to sell out their shares, which is predicted to have
negative impact on firm's performance. Faleye, Mehrota and Morck(2006) holds a even
stronger position against employee ownership whose main rationale is that on employee
ownership there is fundamental discrepancy between the benefits of employees and the
benefits of shareholders. Relative to other firms, labor-controlled firms "deviate more
from value maximization, invest less in long-term assets, take fewer risks, grow more
slowly, create fewer new jobs, and exhibit lower labor and total factor productivity." All of
above theories should be tested empirically.
3. MOTIVATION
The basic purpose of this research is to see whether ESOP in reality plays a positive role
for enhancing corporate performance using a new approach. If the answer is yes, the
following questions should be investigated further: is it the optimal approach to inspire
stuffs' productivities for all kinds of corporations? Which conditions does an effective
ESOP need? Though the relevant researches about the effects of ESOP on corporate
performance are not rare, some limitations existe in most of them. The research goals
can be generalized as follow:
4
3.1 To Compute the Performance of the Firm Concerning Workers'
Efficiency
First of all, the proxy variables for corporate performance are not well-founded in most
of literature. While some researches attach importance on the trend of stock price after
the adoption of ESOP(such as Conte, Blasi, Kruse and Jampani(1996)), others use
financial multiples such as ROE and profit margin for their evaluation(Such as
Borstadtand Zwirlein(1995)). Most of them fail to elucidate why their variable selections
are suitable for evaluating the effects of ESOP adoption on the change of corporate
performance drove by the change of work efficiency of workers. In this research, the
proxy variables would be set precise enough to ensure them reflect the workers'
efficiency of their works exactly.
After the selection of proxy variables, the relevant data concerning firm's efficiency of its
workers will be generalized into a simple score (i.e., a scalar) which reflects the overall
productivity of a specific company. This task can be accomplished within the framework
of Data Envelopment Analysis (DEA) which computes the efficiency of a firm in
transforming multiple-inputs into multiple-outputs in relation to its industrial peers.
Charnes et al.(1978)firstly developed the DEA approach for efficiency analysis and it
has been used widely in evaluating corporate performance. In general, DEA identifies
the most efficient firm with an efficiency score one and provides a measure of
inefficiency for all others with scores between zero and one. So with the help of DEA, it
is easy to figure out which firm in a specific industry performs best/worst with a simple
scalar.
3.2To Develop a Method Calculating the Part of Productivity Change
Caused by the Adoption of ESOP Solely
Secondly, most literatures fail to distinguish the consequences caused by ESOP per se
and the consequences caused by other exogenous factors. For example, even if you see
the corporate financial performance ameliorating along with the adoption of ESOP, you
still can't state that such improvement is definitely the consequence of ESOP because
there is a possibility that it is caused by the improvement of macro-economy rather than
ESOP but these two events concur occasionally. To address this problem, some studies
look at the performance of employee ownership companies compared to that of similar
non-employee ownership companies in order to cancel out their performance change
caused by exogenous environmental impacts. But such approach still implicitly assumes
5
that the effects of environmental impacts on all firms in an industry group are equal
universally, which in reality is not always true.
Graph I is helpful to illustrate the above ideas. From time 1 to time 2, production
possibility frontier (PPF) of a specific industry shifts from the solid curve to the dashed
curve because of the change of some exogenous factors, say the advance of technology
or the improvement of market environment, which can influence all companies in this
industry group. If the input-output situation of a company pertaining such industry shifts
from point A to point B after the adoption of ESOP, can we regard the distance between
A and B just be the improvement of internal endogenous efficiency caused by the
adoption of ESOP of this company? No, because such improvement is caused by both
exogenous (the shift of PPF) and endogenous (the net-movement of input-output point
caused by the change of internal efficiency) factors. At the same time, it is obvious that
the performance change of the firms on/under the upper part of the PPF is caused by
exogenous impacts more than that on/under the lower part of PPF caused by exogenous
impacts because the shift of PPF from time 1 to time 2 is not parallel but has more
movement on its upper part. Thus, the effects of exogenous impacts on the firms
characterized by such PPF are not balanced. In this research, DEA method will be
utilized to overcome these problems and figure out the exact performance change
triggered by ESOP per se.
output
.B
.A
in p u t
Graph I
3.3 To Discover the Crucial Determinants for an Effective ESOP
Another issue concerning employee ownership waiting for thorough investigation is the
search of necessary conditions for a specific ESOP exercising its incentive function
effectively. Some factors should be discriminated as explainable variables for the
performance change caused by ESOP solely.
4. RESEARCH SIGNIFICANCE
The first contribution of this research is to introduce DEA efficiency method into the
6
evaluation of the adoption of ESOP for corporate performance. For the evaluation of
corporate performance change, two common evaluation methods were employed in
existing literature: univariate trend analysis and multivariate regression. However, both
of them are defective because they all not only fail to insulate exogenous impacts from
calculation, but also have difficulty to build a simple but precise index signifying the
change of productivity caused by the adoption of ESOP. But within DEA framework,
these two difficulties can be overcome appropriately.
Secondly, the core determinants affecting the incentive effect of ESOP found in this
research can provide hints about what kinds of companies can gain large benefits in
productivity by adopting ESOP while what kinds of them can't. By employing new
method to evaluate the performance change, some new empirical results may be found
by which we can figure out how employee ownership affects corporate performance
with a new perspective. .
5. METHODOLOGY
Generally, there are four research steps:
I. Select appropriate firms samples from database.
II. Use DEA approach to process the data to form a general score which reflects the
productivity of the company
III. Use DEA approach to measure each firm's productivity change caused solely by the
adoption of ESOP in a specific industry and compare them.
IV. Find crucial variables determining such productivity change. Build regression
model to explain why certain firms' productivities can change more than others'.
They can be explained as follow:
5.1 Ingredient Variables Selection—Step 1
To construct an overall score signifying workers' productivity, its basic ingredients
should be found logically. For example, in a typical firm, workers should combine their
efforts with firm's capital to produce products and create firm's growth opportunities. So
the operating revenue (which signifies the firm's "products") and Tobin Q (which
signifies the firm's "growth opportunities") can be set as two outputs of the firm while
total operating asset (which signifies the firm's capital) and operating cost (which
signifies the workers' effort) can be set as two inputs of the firm. Other logics applied to set
appropriate proxy variables for firm's inputs and outputs can be investigated further.
7
5.2 Measurement of Performance: DEA Method—Step2
Many alternative models for evaluating corporate performance can be applied within the
framework of DEA. According to Cooper et al.(2007), the most basic and simple one
can be laid out as follow:
mi
ì
nu u
,
s.t. u x
o
> X ì
y
o
s Y ì
L s eì s U
ì>0
(1 )
Where x
o
and y
o
are the input and output vectors respectively for the company evaluated
currently (Which is called as Decision Maker Unit, or a DMU); X and Y are the vectors
of the other companies' inputs and outputs vectors forming the production reference
against which the current DMU will be evaluated; e is the unit vector; L and U are two
scalars. The last restricts of this model determines the character of return to scale of all
DMU. A scalaru and a vectorì would be solved from this model. Intuitively, DEA model tries to
find a linear combination of other DMUs (i.e., to find vectorì) which can produce at least
the same amount of the outputs the current DMU can produce while minimizes the
amount of the inputs proportionally the current DMU must have to produce the given
outputs vector. Ifu is solved with number 1, then the current DMU evaluated is performing
efficiently because it is impossible to reduce its inputs anymore while keep its outputs
more or equal to the given level. In other cases where 0<u<1, the
current DMU shows low efficiency.
5.3 Measurement of Performance Change Caused by ESOP—Step 3
For calculating the performance change caused by ESOP solely, the environmental
effects (the shift of PPF) must be insulated from calculation. Fortunately, within the
framework of DEA, it is possible to construct a score only reflecting the internal
endogenous efficiency change. To figure out the internal efficiency change from time sto
time t, two programming problems for a specific company should be solved
respectively:
o
s
= mi
ì
nu u,
s.t. u x
s
o > X
s
ì
y
s
o s Y
s
ì
L s eì s U
ì>0
(2)
o
t
= mi
ì
nu u,
s.t. u x
t
o > X
t
ì
y
t
o s Y
t
ì
L s eì s U
ì>0
(3)
Where s and t are the time superscript. Then the ratio |
t
s
= o
t
/ o
s
can be defined as the
8
internal efficiency improvement/decline index from time s to time t. According to Coelli
et al.(2005), such ratio is just the component of the Malmquist TFP index signifying the
efficiency change with time which can be given by:
|
t
s
= d
o
t ( y
o
t , x
o
t )
s s s
d
o
( y
o
, x
o
) (4)
Where function d denotes the input-distance function for the DMU evaluated at some
specific time point (See detail in Coelli, Rao, Donnell and Battese(2005)).
Graph 2 illustrates the intuitive meaning of the ratio |
t
s
. The solid curve and dashed
curve represent the PPF for an industry in time s and time t respectively. After the
adoption of ESOP at time t, the input-output situation of a particular company
pertaining to this industry shifts from point A to point B with the time from s to t. The
ratio |
t
s
just signifies the ratio of two distance ratios: BE/CE:AF/DF. The amount of this
ratio implies how portion of inputs can be reduced to reach the PPF for the DMU
evaluated at the specific time point. If |
t
s
is calculated larger than 1, we can conclude
that the internal efficiency of the relevant DMU indeed improves from time t to time s.
Else, its internal efficiency declines.
output
.E
.F
.C
. D. A
.B
in p u t
Graph II
After calculating the ratio |
t
s
for all DMUs, the basic hypothesis would be proposed
waiting for test:
Hypothesis 1: After the adoption of ESOP, the internal efficiency of the company would
improve significantly.
Hypothesis 1 can be tested by seeing whether the ratios|
t
s
for the DMUs adopting ESOP
at time t are significantly larger than one. Such tests can take many forms.
9
5.4 Empirical Test and the Factors Determining the Effectiveness of
ESOP—Step 4
A regression model also can be built to test hypothesis 1. Its basic form has the
following specification:
Model1: sgn
i
= o
+
¿
j
|
j
f
ji
+ ¸ ESOP
i
+ c
i
(5)
Where sgn
i
is the sign function of |
t
s,
i
÷1 in which |
t
s,i denotes internal efficiency
improvement/decline index from time t to time s calculated by the above method for the
company i; ESOP
i
is a dummy variable that takes value of one if company i adopted
ESOP at time t or zero otherwise; f
ji
,¬j represent a set of control variables capturing
the characteristics of the firms such as total assets, debt to equity ratio, employees size,
the firm age, etc.; c
i
is a i.i.d. random error; o , |
j
,¸ , ¬j are the parameters to be
calculated. If hypothesis 1 holds, the coefficient of the dummy variable ESOP,|, would
be expected to be positive significantly.
The next goal is to find the core determinants for the effectiveness of ESOP improving
corporate performance. To this end, another linear regression model will be built whose
basic form can be specified as follow:
Model2 : |
t
s,i
ESOP=1
= a
+
¿
j
b
j
v
ji +
¿
j
c
j
f
ji
+ e
i
( 6)
Where |
t
s,i is the internal efficiency improvement/decline index from time t to
ESOP
i
=1
time s for the company i adopting ESOP at time t; v
ji
, ¬j represent a set of variables
capturing the characteristics of the firms concerning the effectiveness of an ESOP;
f
ji
,¬j still represent a set of control variables; e
i
is another i.i.d. random error;
a , b
j
, c
j
, ¬j are the parameters to be calculated. According to existing literature and
experience, the candidates of v
j
can be:
? The proportion of employee ownership accounting for the whole shares. It is
intuitive to hypothesize that with the increase of such proportion, the productivity of
the firm should improve because workers have more incentive to behave well for
their stakes.
? The employees' control power and participation. Weston, Mitchell and
Mulherin(2004) points out that through ESOP only if the workers indeed are given
chance to participate in decision-making process in their role of shareholders can
their productivity improve. The problem is that management has been usually
10
unwilling to grant workers full shareholder rights. However, a opposite viewpoint
about worker's participation in decision-making is proposed by Faleye, Mehrota and
Morck(2006), which suggests that labor-controlled firms deviate more from value
maximization . Provided with control rights, workers prefer to pursue more stable
and higher wage rather than improving the value of the firm in the long-term. Thus,
how workers participation in decision-making effects corporate performance should
be investigated empirically. An example of the proxy variable for workers control
power is the portion of shares having voting rights pertaining the employee
ownership.
? The volatility of the stock price. According to normal linear contracting agency
model proposed by Holmstrom and Milgrom(1991) as well as Holmstrom and
Milgrom(1987), in the optimal contract the employees' shares should be higher to
induce workers' more efforts when the variance of firm value is smaller. So we can
hypothesize that holding the employees' shares unchanged, more variance of firm
value leads to lower incentive effect of the ESOP. The volatility of the stock price
can be set as the proxy variable for firm value variance put into the regression
model to see whether such hypothesis is true empirically.
? The direct purpose of the ESOP. Though Borstadt and Zwirlein(1995) find no
evidences that the direct reasons for ESOP adoption (according to Borstadt and
Zwirlein(1995), typically they are defense takeover, wage concession, and
motivating tool respectively) are significant variables leading corporate
productivity change, we still set them as a explainable variable for our regression
model because we want to see whether the different measurement of performance
change in our research can render us new empirical results. .
? The industry classification. For every single industry, the value of its workers
diverges greatly. Thus, we propose that in different industry, workers would react to
employee ownership in different way.
? The employees average wage. Klein and Hall(1988) points out that one key
individual level factor determining the satisfactory towards ESOP is the employee'
status within the plan. Blasi, Kruse and Markowiz(2008) find that workers'
economic insecurity can diminish the incentive effect of the ownership they own. If
the workers have low income, they may fail to resist financial risk concerning their
ownership, which depress their motivation to work hard. So we set employees
average wage as another independent variable in model 2.
? The corporate culture. Last but not least, corporate culture can make difference for
the success of a specific ESOP. Blasi, Kruse and Markowiz(2008) proposes that a
worker's economic insecurity and response to "shared capitalism"(which refers to the
ownership shared by all stuffs in the company) are likely to be related to worker
empowerment and perceived fairness. Good employee-management relationship
may be an important condition to create cooperation and higher performance under
11
"shared capitalism" ownership where "psychological contract" between employees
and the firm is tight and positive (Rousseau and Shperling(2003)). Unfortunately,
the proxy variable for corporate culture could be too ambiguous to be useful. The
workers turnover rate can be a potential proxy variable for the "goodness" of
corporate culture because we can propose logically that good work atmosphere
would dampen workers' desire to change their jobs.
The focus at this point is to test the sign of every b
j
, ¬j statistically based on the
hypothesis proposed.
6. DATA
The research samples would be selected from the firms of USA after 1976 when the
federal government established Employee Retirement Income Security Act (ERISA).
Data can be gathered from many databases such as COMPUSTAT and Wall Street
Journal. Some special research organizations such as National Center for Employee
Ownership (NCEO) also provide useful data and information about historical ESOPs.
The one-hand financial information about the firm can be obtained from their annual
reports and proxy statements. Generally speak, relevant data are not scarce. The detailed
method and scope of data selection would be refined when the research is undertaken.
7. SUMMARY: RESEARCH PROCESS
The following illustrative graph can summarize the whole process of this research:
Data Selection
Model 1
?
ts
DEA Build Regression Model
Model 2
Graph 3
12
References:
Aoki, M. The Cooperative Game Theory of the Firm. England: Oxford University
Press, 1984.
Beatty, A. The Cash Flow and Informational Effects of Employee Stock Ownership
Plans. Journal of Financial Economics, 1995, (38), pp. 211-40.
Borstadt, Lisa F and Zwirlein, Thomas J. ESOPs in Publicly Held Companies:
Evidence On Productivity and Firm Performance. Journal of Financial and Strategic
Decisions, 1995, 8(1), pp. 1-13.
Blasi, Joseph R; Kruse, Douglas L and Markowiz, Harry M. Risk and Lack of
Diversification Under Employee Ownership and Shared Capitalism,NBER , 2008.
Charnes, A.; Cooper, W. W. and Rhodes, E. Measuring the Efficiency of Decision
Making Units. European Journal of Operational Research , 1978, 2, pp. 429-44.
Coelli, Timothy, J; Rao, D S Prasada; Donnell, Christopher J O' and Battese, George
E. An Introduction to Efficiency and Productivity Analysis. Springer Science+Business
Media, 2005.
Conte, Michael A.; Blasi, Joseph; Kruse, Douglas and Jampani, Rama. Financial
Returns of Public Esop Companies: Investor Effects Vs. Manager Effects. Financial
Analysts Journal, 1996, 52(4), pp. 51-61.
Cooper, William W.; Seiford, Lawrence M. and Tone, Kaoru. Data Envelopment
Analysis:a Comprehensive Text with Models, Applications, References and Dea-Solver
Software. Springer Science+Business Media, 2007.
Drucker, P. F. The Future of Industrial Man: A Conservative Approach. New York:
The John Day Company, 1978.
Faleye, Olubunmi; Mehrota, Vikas and Morck, Randall. When Labor Has a Vocie in
Corporate Governance. Journal of Financial and Quantitative Analysis, 2006, 41(3), pp.
489-510.
Henry, Steve; Kavanaugh, Joseph; Strecher, Robert and Chisholm, Darla. Esop Firm
Performance Pre- And Post-Market Peak: Empirical Evidence. Academy of Accounting
and Financial Studies Journal, 2007, 11(1).
Holmstrom, B and Milgrom, P R. Aggregation and Linearity in the Provision of
Intertemporal Incentives. Econometrica, 1987, 55, pp. 308-28.
Holmstrom, B. and Milgrom, P. R. Multi-Task Principal-Agent Analyses: Incentive
Contracts, Asset Ownership and Job Design. Journal of Law, Economics and
Organization, 1991, 7, pp. 524-52.
Kato, Takao and Jones, Derek C. The Productivity Effects of Employee
Stock-Ownership Plans and Bonuses: Evidence Fromjapanese Panel Data. The
American Economic Review, 1995, 85(3), pp. 391-414.
13
Kim, E. Han. Employee Capitalism Or Corporate Socialism? Broad-Based Employee
Stock Ownership,American Finance Association 2010 Annual Conference. , 2010.
Klein, Katharine J. and Hall, Rosalie J. Correlates of Employee Satisfaction with
Stock Ownership:Who Likes an Esop Most?. Journal of Applied Psychology, 1988,
73(4), pp. 630-38.
Kruse, D. Profit Sharing and Productivity: Microeconomic Evidence From the United
States. Economic Journal, 1992, 102(410), pp. 24-36.
Muelbroek, Lisa. Company Stock in Pension Plans: How Costly is It?,Harvard
Business School , 2002, 2-25.
Pagano, Marco and Volpin, Paolo. Managers, Workers, and Corporate Control. The
Journal of Finance, 2003, 60(2), pp. 841-68.
Park, S. and Song, M. H. Employee Stock Ownership Plans, Firm Performance, and
Monitoring by Outside Blockholders. Financial Management, 1995, 24(4), pp. 52-65.
Pugha, William N.; Oswald, Sharon L. and Jr. Jahera, John S. The Effect of Esop
Adoptions On Corporate Performance: Are there Really Performance Changes?.
Managerial and Decision Economics, 2000, (21), pp. 167-80.
Rousseau, Denise M. and Shperling, Zipi. The, Pieces of the Action Relationship,
Changing Employment. Academy of Management Review, 2003, 28(4), pp. 553-70.
Vanek, J. General Equilibrium of International Discrimination: The Case of Customs
Union. Cambrige: Harvard University Press, 1965.
Weston, Fred; Mitchell, Mark and Mulherin, Harold. Takeovers, Restructuring and
Corporate Governance. Pearson/Prentice Hall, 2004.
14
doc_520049927.docx
Research Proposal On The Incentive Effect Of Employee Ownership: Another Measurement And Its Determinant:- An employee stock ownership plan (ESOP)[1] is an employee-owner scheme that provides a company's workforce with an ownership interest in the company. In an ESOP, companies provide their employees with stock ownership, often at no up-front cost to the employees.
Research Proposal On The Incentive Effect Of
Employee Ownership: Another Measurement And
Its Determinants
Abstract:
Noticing the limitations in most of relevant empirical studies, this research tries to
develop a new method to (1): test whether the corporate performance change led by the
adoption of ESOP per se is significant, and (2): figure out the core determinants for the
incentive effect of equity ownership. After appropriate selection of proxy variables for
signifying firms' inputs and outputs, DEA approach will be employed to evaluate firm's
efficiency change. Then the regression models will be built to achieve the above two
goals.
1
Contents
1.BACKGROUND.................................................................................................................................... 3
2.LITERATURE REVIEW....................................................................................................................... 3-4
3.MOTIVATION................................................................................................................................... 4-6
3.1To Compute the Performance of the firm concerning Workers
Efficiency................5
3.2 To Develop a Method Calculating the Part of Productivity Change Caused
By the Adoption of ESOP
Solely.....................................5-6
3.1To Discover the Crucial Determinants for an Effective
ESOP......................6
4.RESEARCH SIGNIFICANCE................................................................................................................6-7
5.METHODOLOGY............................................................................................................................ 7-12
5.1 Ingredient Variables Selection—Step
1....................................7 5.2 Measurement of
Performance: DEA Method—Step2........................7-8 5.3
Measurement of Performance Change Caused by ESOP—Step
3................8-9 5.4Empirical Test and the Factors Determining the
Effectiveness of ESOP—Step 4......10-12
6.DATA................................................................................................................................................. 12
7.REFERENCES................................................................................................................................13-14
2
1. BACKGROUND
Employee stock ownership plans (ESOPs) have received much attention in the popular
business press as well as the academic field of corporate finance (Borstadt and
Zwirlein(1995)). In a specific ESOP, non-executive employees are given chances to own
stock, which makes it possible for them to hold some residual claims on the company
like shareholders, thereby the so-called "people's capitalism" is expected to be achieved.
The supporters of ESOPs usually stress its incentive effect on workers, whose argument
is that through ESOP participation, employees will be more motivated and begin to
think and act as owners. They will exercise fiduciary-like responsibility over the
resources of the firm, minimize agency costs, and align their behavior with the goals of
the firm (Henry et al.(2007)). If this is true, it is reasonable to anticipate that an ESOP-
based firm would outperform comparable firms in its industry group. Such hypothesis
has been supported by many empirical works concerning the firms all around world. But
there are also opposite viewpoints against the incentive hypothesis of ESOPs. Using
different samples or statistical methods, some empirical evidences also cast doubts on
the validity of the proposition that the adoption of ESOPs can improve corporate
performance significantly.
Without consistent empirical results, how employee ownership actually affects workers'
productivity and corporate performance is still an open question. Whether and under
what conditions such practices make sense or are really ill-advised should be found by
appropriate approach. This research tries to develop a method to investigate this old
question with a new perspective.
2. LITERATURE REVIEW
The basic rationale for employee ownership is that it leads to the convergence of
interests employees' interests with those of shareholders, which inspire workers to work
hard for achieving better corporate performance (Vanek(1965), Drucker(1978) and
Aoki(1984)). However, empirical results about the incentive hypothesis of employee
ownership are ambiguous. A detailed studies review about employee ownership and
corporate performance can be found in The National Center for Employee Ownership
(NCEO)
1
. While some studies show positive relationship between them (such as
Kruse(1992), Kato and Jones(1995)and Kim(2010)), there are also studies render an
opposite conclusion(such as Faleye et al.(2006)). But general speaking, most of
1
"Research on Employee Ownership, Corporate Performance and Employee Compensation",
http://www.nceo.org/main/article.php/id/3/
3
empirical results provide mixed evidences (Weston et al.(2004)). For example, Park and
Song(1995) finds that though firms with outside block holders adopting ESOP
experience a permanent improvement in performance, ESOP makes the firms'
performance without block holders declines. Conte et al.(1996)finds that financial
returns of public companies with ESOP are higher than those of comparable non-ESOP
companies but decline after the adoption of ESOP. Borstadt and Zwirlein(1995)finds
some evidences that ESOPs established as an incremental employee benefits results in
productivity improvement. However, after adjustment for industry data, no significant
change of performance is observed. Pugha et al.(2000)finds that few of financial
performance do improve significantly after the adoption of ESOP, but this appears to be
largely short-term.
The inconsistency of empirical results also leads scholars to cast doubts on the validity
of incentive effect of employee ownership. Some other theories about the consequence
of employee ownership have been proposed. In the view of investment and portfolio
selection, Muelbroek(2002) argue that any type of employee ownership all violates the
basic principle of portfolio diversification, which would make workers worse-off. From
this sense, ESOPs will not form an effective stimulus to inspire workers but just expose
them to unnecessary financial risks. Beatty(1995), Pagano and Volpin(2003)constructs
another counterargument against employee ownership that the adoption of ESOP
usually serves as a determent to hostile takeover because workers are stable
shareholders who would not be likely to sell out their shares, which is predicted to have
negative impact on firm's performance. Faleye, Mehrota and Morck(2006) holds a even
stronger position against employee ownership whose main rationale is that on employee
ownership there is fundamental discrepancy between the benefits of employees and the
benefits of shareholders. Relative to other firms, labor-controlled firms "deviate more
from value maximization, invest less in long-term assets, take fewer risks, grow more
slowly, create fewer new jobs, and exhibit lower labor and total factor productivity." All of
above theories should be tested empirically.
3. MOTIVATION
The basic purpose of this research is to see whether ESOP in reality plays a positive role
for enhancing corporate performance using a new approach. If the answer is yes, the
following questions should be investigated further: is it the optimal approach to inspire
stuffs' productivities for all kinds of corporations? Which conditions does an effective
ESOP need? Though the relevant researches about the effects of ESOP on corporate
performance are not rare, some limitations existe in most of them. The research goals
can be generalized as follow:
4
3.1 To Compute the Performance of the Firm Concerning Workers'
Efficiency
First of all, the proxy variables for corporate performance are not well-founded in most
of literature. While some researches attach importance on the trend of stock price after
the adoption of ESOP(such as Conte, Blasi, Kruse and Jampani(1996)), others use
financial multiples such as ROE and profit margin for their evaluation(Such as
Borstadtand Zwirlein(1995)). Most of them fail to elucidate why their variable selections
are suitable for evaluating the effects of ESOP adoption on the change of corporate
performance drove by the change of work efficiency of workers. In this research, the
proxy variables would be set precise enough to ensure them reflect the workers'
efficiency of their works exactly.
After the selection of proxy variables, the relevant data concerning firm's efficiency of its
workers will be generalized into a simple score (i.e., a scalar) which reflects the overall
productivity of a specific company. This task can be accomplished within the framework
of Data Envelopment Analysis (DEA) which computes the efficiency of a firm in
transforming multiple-inputs into multiple-outputs in relation to its industrial peers.
Charnes et al.(1978)firstly developed the DEA approach for efficiency analysis and it
has been used widely in evaluating corporate performance. In general, DEA identifies
the most efficient firm with an efficiency score one and provides a measure of
inefficiency for all others with scores between zero and one. So with the help of DEA, it
is easy to figure out which firm in a specific industry performs best/worst with a simple
scalar.
3.2To Develop a Method Calculating the Part of Productivity Change
Caused by the Adoption of ESOP Solely
Secondly, most literatures fail to distinguish the consequences caused by ESOP per se
and the consequences caused by other exogenous factors. For example, even if you see
the corporate financial performance ameliorating along with the adoption of ESOP, you
still can't state that such improvement is definitely the consequence of ESOP because
there is a possibility that it is caused by the improvement of macro-economy rather than
ESOP but these two events concur occasionally. To address this problem, some studies
look at the performance of employee ownership companies compared to that of similar
non-employee ownership companies in order to cancel out their performance change
caused by exogenous environmental impacts. But such approach still implicitly assumes
5
that the effects of environmental impacts on all firms in an industry group are equal
universally, which in reality is not always true.
Graph I is helpful to illustrate the above ideas. From time 1 to time 2, production
possibility frontier (PPF) of a specific industry shifts from the solid curve to the dashed
curve because of the change of some exogenous factors, say the advance of technology
or the improvement of market environment, which can influence all companies in this
industry group. If the input-output situation of a company pertaining such industry shifts
from point A to point B after the adoption of ESOP, can we regard the distance between
A and B just be the improvement of internal endogenous efficiency caused by the
adoption of ESOP of this company? No, because such improvement is caused by both
exogenous (the shift of PPF) and endogenous (the net-movement of input-output point
caused by the change of internal efficiency) factors. At the same time, it is obvious that
the performance change of the firms on/under the upper part of the PPF is caused by
exogenous impacts more than that on/under the lower part of PPF caused by exogenous
impacts because the shift of PPF from time 1 to time 2 is not parallel but has more
movement on its upper part. Thus, the effects of exogenous impacts on the firms
characterized by such PPF are not balanced. In this research, DEA method will be
utilized to overcome these problems and figure out the exact performance change
triggered by ESOP per se.
output
.B
.A
in p u t
Graph I
3.3 To Discover the Crucial Determinants for an Effective ESOP
Another issue concerning employee ownership waiting for thorough investigation is the
search of necessary conditions for a specific ESOP exercising its incentive function
effectively. Some factors should be discriminated as explainable variables for the
performance change caused by ESOP solely.
4. RESEARCH SIGNIFICANCE
The first contribution of this research is to introduce DEA efficiency method into the
6
evaluation of the adoption of ESOP for corporate performance. For the evaluation of
corporate performance change, two common evaluation methods were employed in
existing literature: univariate trend analysis and multivariate regression. However, both
of them are defective because they all not only fail to insulate exogenous impacts from
calculation, but also have difficulty to build a simple but precise index signifying the
change of productivity caused by the adoption of ESOP. But within DEA framework,
these two difficulties can be overcome appropriately.
Secondly, the core determinants affecting the incentive effect of ESOP found in this
research can provide hints about what kinds of companies can gain large benefits in
productivity by adopting ESOP while what kinds of them can't. By employing new
method to evaluate the performance change, some new empirical results may be found
by which we can figure out how employee ownership affects corporate performance
with a new perspective. .
5. METHODOLOGY
Generally, there are four research steps:
I. Select appropriate firms samples from database.
II. Use DEA approach to process the data to form a general score which reflects the
productivity of the company
III. Use DEA approach to measure each firm's productivity change caused solely by the
adoption of ESOP in a specific industry and compare them.
IV. Find crucial variables determining such productivity change. Build regression
model to explain why certain firms' productivities can change more than others'.
They can be explained as follow:
5.1 Ingredient Variables Selection—Step 1
To construct an overall score signifying workers' productivity, its basic ingredients
should be found logically. For example, in a typical firm, workers should combine their
efforts with firm's capital to produce products and create firm's growth opportunities. So
the operating revenue (which signifies the firm's "products") and Tobin Q (which
signifies the firm's "growth opportunities") can be set as two outputs of the firm while
total operating asset (which signifies the firm's capital) and operating cost (which
signifies the workers' effort) can be set as two inputs of the firm. Other logics applied to set
appropriate proxy variables for firm's inputs and outputs can be investigated further.
7
5.2 Measurement of Performance: DEA Method—Step2
Many alternative models for evaluating corporate performance can be applied within the
framework of DEA. According to Cooper et al.(2007), the most basic and simple one
can be laid out as follow:
mi
ì
nu u
,
s.t. u x
o
> X ì
y
o
s Y ì
L s eì s U
ì>0
(1 )
Where x
o
and y
o
are the input and output vectors respectively for the company evaluated
currently (Which is called as Decision Maker Unit, or a DMU); X and Y are the vectors
of the other companies' inputs and outputs vectors forming the production reference
against which the current DMU will be evaluated; e is the unit vector; L and U are two
scalars. The last restricts of this model determines the character of return to scale of all
DMU. A scalaru and a vectorì would be solved from this model. Intuitively, DEA model tries to
find a linear combination of other DMUs (i.e., to find vectorì) which can produce at least
the same amount of the outputs the current DMU can produce while minimizes the
amount of the inputs proportionally the current DMU must have to produce the given
outputs vector. Ifu is solved with number 1, then the current DMU evaluated is performing
efficiently because it is impossible to reduce its inputs anymore while keep its outputs
more or equal to the given level. In other cases where 0<u<1, the
current DMU shows low efficiency.
5.3 Measurement of Performance Change Caused by ESOP—Step 3
For calculating the performance change caused by ESOP solely, the environmental
effects (the shift of PPF) must be insulated from calculation. Fortunately, within the
framework of DEA, it is possible to construct a score only reflecting the internal
endogenous efficiency change. To figure out the internal efficiency change from time sto
time t, two programming problems for a specific company should be solved
respectively:
o
s
= mi
ì
nu u,
s.t. u x
s
o > X
s
ì
y
s
o s Y
s
ì
L s eì s U
ì>0
(2)
o
t
= mi
ì
nu u,
s.t. u x
t
o > X
t
ì
y
t
o s Y
t
ì
L s eì s U
ì>0
(3)
Where s and t are the time superscript. Then the ratio |
t
s
= o
t
/ o
s
can be defined as the
8
internal efficiency improvement/decline index from time s to time t. According to Coelli
et al.(2005), such ratio is just the component of the Malmquist TFP index signifying the
efficiency change with time which can be given by:
|
t
s
= d
o
t ( y
o
t , x
o
t )
s s s
d
o
( y
o
, x
o
) (4)
Where function d denotes the input-distance function for the DMU evaluated at some
specific time point (See detail in Coelli, Rao, Donnell and Battese(2005)).
Graph 2 illustrates the intuitive meaning of the ratio |
t
s
. The solid curve and dashed
curve represent the PPF for an industry in time s and time t respectively. After the
adoption of ESOP at time t, the input-output situation of a particular company
pertaining to this industry shifts from point A to point B with the time from s to t. The
ratio |
t
s
just signifies the ratio of two distance ratios: BE/CE:AF/DF. The amount of this
ratio implies how portion of inputs can be reduced to reach the PPF for the DMU
evaluated at the specific time point. If |
t
s
is calculated larger than 1, we can conclude
that the internal efficiency of the relevant DMU indeed improves from time t to time s.
Else, its internal efficiency declines.
output
.E
.F
.C
. D. A
.B
in p u t
Graph II
After calculating the ratio |
t
s
for all DMUs, the basic hypothesis would be proposed
waiting for test:
Hypothesis 1: After the adoption of ESOP, the internal efficiency of the company would
improve significantly.
Hypothesis 1 can be tested by seeing whether the ratios|
t
s
for the DMUs adopting ESOP
at time t are significantly larger than one. Such tests can take many forms.
9
5.4 Empirical Test and the Factors Determining the Effectiveness of
ESOP—Step 4
A regression model also can be built to test hypothesis 1. Its basic form has the
following specification:
Model1: sgn
i
= o
+
¿
j
|
j
f
ji
+ ¸ ESOP
i
+ c
i
(5)
Where sgn
i
is the sign function of |
t
s,
i
÷1 in which |
t
s,i denotes internal efficiency
improvement/decline index from time t to time s calculated by the above method for the
company i; ESOP
i
is a dummy variable that takes value of one if company i adopted
ESOP at time t or zero otherwise; f
ji
,¬j represent a set of control variables capturing
the characteristics of the firms such as total assets, debt to equity ratio, employees size,
the firm age, etc.; c
i
is a i.i.d. random error; o , |
j
,¸ , ¬j are the parameters to be
calculated. If hypothesis 1 holds, the coefficient of the dummy variable ESOP,|, would
be expected to be positive significantly.
The next goal is to find the core determinants for the effectiveness of ESOP improving
corporate performance. To this end, another linear regression model will be built whose
basic form can be specified as follow:
Model2 : |
t
s,i
ESOP=1
= a
+
¿
j
b
j
v
ji +
¿
j
c
j
f
ji
+ e
i
( 6)
Where |
t
s,i is the internal efficiency improvement/decline index from time t to
ESOP
i
=1
time s for the company i adopting ESOP at time t; v
ji
, ¬j represent a set of variables
capturing the characteristics of the firms concerning the effectiveness of an ESOP;
f
ji
,¬j still represent a set of control variables; e
i
is another i.i.d. random error;
a , b
j
, c
j
, ¬j are the parameters to be calculated. According to existing literature and
experience, the candidates of v
j
can be:
? The proportion of employee ownership accounting for the whole shares. It is
intuitive to hypothesize that with the increase of such proportion, the productivity of
the firm should improve because workers have more incentive to behave well for
their stakes.
? The employees' control power and participation. Weston, Mitchell and
Mulherin(2004) points out that through ESOP only if the workers indeed are given
chance to participate in decision-making process in their role of shareholders can
their productivity improve. The problem is that management has been usually
10
unwilling to grant workers full shareholder rights. However, a opposite viewpoint
about worker's participation in decision-making is proposed by Faleye, Mehrota and
Morck(2006), which suggests that labor-controlled firms deviate more from value
maximization . Provided with control rights, workers prefer to pursue more stable
and higher wage rather than improving the value of the firm in the long-term. Thus,
how workers participation in decision-making effects corporate performance should
be investigated empirically. An example of the proxy variable for workers control
power is the portion of shares having voting rights pertaining the employee
ownership.
? The volatility of the stock price. According to normal linear contracting agency
model proposed by Holmstrom and Milgrom(1991) as well as Holmstrom and
Milgrom(1987), in the optimal contract the employees' shares should be higher to
induce workers' more efforts when the variance of firm value is smaller. So we can
hypothesize that holding the employees' shares unchanged, more variance of firm
value leads to lower incentive effect of the ESOP. The volatility of the stock price
can be set as the proxy variable for firm value variance put into the regression
model to see whether such hypothesis is true empirically.
? The direct purpose of the ESOP. Though Borstadt and Zwirlein(1995) find no
evidences that the direct reasons for ESOP adoption (according to Borstadt and
Zwirlein(1995), typically they are defense takeover, wage concession, and
motivating tool respectively) are significant variables leading corporate
productivity change, we still set them as a explainable variable for our regression
model because we want to see whether the different measurement of performance
change in our research can render us new empirical results. .
? The industry classification. For every single industry, the value of its workers
diverges greatly. Thus, we propose that in different industry, workers would react to
employee ownership in different way.
? The employees average wage. Klein and Hall(1988) points out that one key
individual level factor determining the satisfactory towards ESOP is the employee'
status within the plan. Blasi, Kruse and Markowiz(2008) find that workers'
economic insecurity can diminish the incentive effect of the ownership they own. If
the workers have low income, they may fail to resist financial risk concerning their
ownership, which depress their motivation to work hard. So we set employees
average wage as another independent variable in model 2.
? The corporate culture. Last but not least, corporate culture can make difference for
the success of a specific ESOP. Blasi, Kruse and Markowiz(2008) proposes that a
worker's economic insecurity and response to "shared capitalism"(which refers to the
ownership shared by all stuffs in the company) are likely to be related to worker
empowerment and perceived fairness. Good employee-management relationship
may be an important condition to create cooperation and higher performance under
11
"shared capitalism" ownership where "psychological contract" between employees
and the firm is tight and positive (Rousseau and Shperling(2003)). Unfortunately,
the proxy variable for corporate culture could be too ambiguous to be useful. The
workers turnover rate can be a potential proxy variable for the "goodness" of
corporate culture because we can propose logically that good work atmosphere
would dampen workers' desire to change their jobs.
The focus at this point is to test the sign of every b
j
, ¬j statistically based on the
hypothesis proposed.
6. DATA
The research samples would be selected from the firms of USA after 1976 when the
federal government established Employee Retirement Income Security Act (ERISA).
Data can be gathered from many databases such as COMPUSTAT and Wall Street
Journal. Some special research organizations such as National Center for Employee
Ownership (NCEO) also provide useful data and information about historical ESOPs.
The one-hand financial information about the firm can be obtained from their annual
reports and proxy statements. Generally speak, relevant data are not scarce. The detailed
method and scope of data selection would be refined when the research is undertaken.
7. SUMMARY: RESEARCH PROCESS
The following illustrative graph can summarize the whole process of this research:
Data Selection
Model 1
?
ts
DEA Build Regression Model
Model 2
Graph 3
12
References:
Aoki, M. The Cooperative Game Theory of the Firm. England: Oxford University
Press, 1984.
Beatty, A. The Cash Flow and Informational Effects of Employee Stock Ownership
Plans. Journal of Financial Economics, 1995, (38), pp. 211-40.
Borstadt, Lisa F and Zwirlein, Thomas J. ESOPs in Publicly Held Companies:
Evidence On Productivity and Firm Performance. Journal of Financial and Strategic
Decisions, 1995, 8(1), pp. 1-13.
Blasi, Joseph R; Kruse, Douglas L and Markowiz, Harry M. Risk and Lack of
Diversification Under Employee Ownership and Shared Capitalism,NBER , 2008.
Charnes, A.; Cooper, W. W. and Rhodes, E. Measuring the Efficiency of Decision
Making Units. European Journal of Operational Research , 1978, 2, pp. 429-44.
Coelli, Timothy, J; Rao, D S Prasada; Donnell, Christopher J O' and Battese, George
E. An Introduction to Efficiency and Productivity Analysis. Springer Science+Business
Media, 2005.
Conte, Michael A.; Blasi, Joseph; Kruse, Douglas and Jampani, Rama. Financial
Returns of Public Esop Companies: Investor Effects Vs. Manager Effects. Financial
Analysts Journal, 1996, 52(4), pp. 51-61.
Cooper, William W.; Seiford, Lawrence M. and Tone, Kaoru. Data Envelopment
Analysis:a Comprehensive Text with Models, Applications, References and Dea-Solver
Software. Springer Science+Business Media, 2007.
Drucker, P. F. The Future of Industrial Man: A Conservative Approach. New York:
The John Day Company, 1978.
Faleye, Olubunmi; Mehrota, Vikas and Morck, Randall. When Labor Has a Vocie in
Corporate Governance. Journal of Financial and Quantitative Analysis, 2006, 41(3), pp.
489-510.
Henry, Steve; Kavanaugh, Joseph; Strecher, Robert and Chisholm, Darla. Esop Firm
Performance Pre- And Post-Market Peak: Empirical Evidence. Academy of Accounting
and Financial Studies Journal, 2007, 11(1).
Holmstrom, B and Milgrom, P R. Aggregation and Linearity in the Provision of
Intertemporal Incentives. Econometrica, 1987, 55, pp. 308-28.
Holmstrom, B. and Milgrom, P. R. Multi-Task Principal-Agent Analyses: Incentive
Contracts, Asset Ownership and Job Design. Journal of Law, Economics and
Organization, 1991, 7, pp. 524-52.
Kato, Takao and Jones, Derek C. The Productivity Effects of Employee
Stock-Ownership Plans and Bonuses: Evidence Fromjapanese Panel Data. The
American Economic Review, 1995, 85(3), pp. 391-414.
13
Kim, E. Han. Employee Capitalism Or Corporate Socialism? Broad-Based Employee
Stock Ownership,American Finance Association 2010 Annual Conference. , 2010.
Klein, Katharine J. and Hall, Rosalie J. Correlates of Employee Satisfaction with
Stock Ownership:Who Likes an Esop Most?. Journal of Applied Psychology, 1988,
73(4), pp. 630-38.
Kruse, D. Profit Sharing and Productivity: Microeconomic Evidence From the United
States. Economic Journal, 1992, 102(410), pp. 24-36.
Muelbroek, Lisa. Company Stock in Pension Plans: How Costly is It?,Harvard
Business School , 2002, 2-25.
Pagano, Marco and Volpin, Paolo. Managers, Workers, and Corporate Control. The
Journal of Finance, 2003, 60(2), pp. 841-68.
Park, S. and Song, M. H. Employee Stock Ownership Plans, Firm Performance, and
Monitoring by Outside Blockholders. Financial Management, 1995, 24(4), pp. 52-65.
Pugha, William N.; Oswald, Sharon L. and Jr. Jahera, John S. The Effect of Esop
Adoptions On Corporate Performance: Are there Really Performance Changes?.
Managerial and Decision Economics, 2000, (21), pp. 167-80.
Rousseau, Denise M. and Shperling, Zipi. The, Pieces of the Action Relationship,
Changing Employment. Academy of Management Review, 2003, 28(4), pp. 553-70.
Vanek, J. General Equilibrium of International Discrimination: The Case of Customs
Union. Cambrige: Harvard University Press, 1965.
Weston, Fred; Mitchell, Mark and Mulherin, Harold. Takeovers, Restructuring and
Corporate Governance. Pearson/Prentice Hall, 2004.
14
doc_520049927.docx