Asset Sharing and Stakeholder Arrangements:
Human Capital Investments, the Distribution of Powers, and
The Role of Property Rights and Economic Institutions
by
J. Jerome Hughes
B.A. Sociology
University of Alabama at Birmingham, 2000
ABRGJEF
Ed.M/J.D, Harvard University, 2003/2005
SUBMITTED TO THE SLOAN SCHOOL OF MANAGEMENT IN PARTIAL
FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE IN MANAGEMENT RESEARCH
AT THE
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
FEBRUARY 2014
© 2014 Massachusetts Institute of Technology. All rights reserved.
Signature of Author:
Sloail'School of Management
danuary 15, 2014
Certified by:
Thomas Anton Kochan
George Maverick Bunker Professor of Management
Thesis Supervisor
Accepted by:
Ezra Zuckerman Sivan
University Professor
Technological
Nanyang
Professor of Technological Innovation, Entrepreneurship, and Strategic Management
Chair MIT Sloan PhD Program
Asset Sharing and Stakeholder Arrangements:
Human Capital Investments, the Distribution of Powers, and
The Role of Property Rights and Economic Institutions
by
J. Jerome Hughes
Submitted to the Sloan School of Management
on January 16, 2014 in Partial Fulfillment of the
Requirements for the Degree of Master of Science in
Management Research
ABSTRACT
Theories of human capital investment, which emphasize encouraging and protecting investments in
human capital, have become salient in rationalizing the adoption of firm asset-sharing and employee
stakeholder arrangements, such as Employee Stock Ownership Plans. Yet, mechanisms such as
bargaining power have also been a key part of the literature on employee and firm bargaining outcomes.
Part of the puzzle with bargaining power as an explanation is that not all forms of bargaining power
significantly explain the adoption of firm asset sharing and employee stakeholder arrangements. In order
to provide an improved explanation for the adoption of these arrangements, we utilize distributive
conceptions of property rights and economic institutions to highlight how power is allocated, segmented,
and distributed by economic institutions and, thus, impacts firm asset sharing and employee stakeholder
arrangements.
Thesis Supervisor: Thomas Anton Kochan
Title: George Maverick Bunker Professor of Management
0.1 Introduction
Theories of human capital and specific investment have become salient in rationalizing
why firms adopt asset sharing and employee stakeholding arrangements, such as broad-based
stock ownership plans (Blair, 1995; Robinson and Zhang, 2005; Pendleton and Robinson, 2011).
While scholars have documented the impact of these arrangements on employee outcomes,
corporate performance, and organizational stability and growth, more recent theories have sought
to explain the rationale behind these arrangements from a different perspective (Kaarsemaker, et
al. 2009; Freeman et. al, 2010; Carberry, 2011).
These recent arguments propose that
incentivizing and protecting investments in human capital are key impetuses for the adoption of
these arrangements by firms (Blair, 1999; Pendleton and Robinson, 2011; Robinson and Zhang,
2005; Roberts and Van de Steen, 2000; Inderst and Mueller, 2007). These investments may be
key to improving individual worker performance, teamwork and product quality (Kochan, Katz,
and McKersie, 1986).
Yet, scholars have also argued that other non-efficiency oriented mechanisms, such as
bargaining power, determine the interaction between firms and individuals, and thus, shape
outcomes related to wages, benefits, and job quality (Osternan, 1999). In fact, in the same
human capital investment literature, which relies on theories of incomplete contracts, the
dynamic of bargaining power is essential to establishing wages and benefits and amount of firm
assets that are shared with employees (Aoki, 1980, 1984; Grout, 1984; Moretto and Rossini,
1995, 1996; Jou and Lee; 2004). In management theory, labor stakeholder theorists have argued,
among other things, that bargaining power is a key factor in the adoption of labor stakeholder
and firm-asset sharing arrangements (Kochan and Rubenstein, 2000; Coff, 1999). But, while the
literature predicts that bargaining power should be a significant factor in the adoption of these
3 Asset Sharing and Stakeholder Arrangements
arrangements, an empirical challenge has emerged.
The literature on Employee Stock
Ownership Plans, or ESOPs, indicates that in the early years, unions did not significantly
contribute to the adoption of these arrangements (McElrath and Rowan, 1992; McHugh,
Cutcher-Gershenfeld, and Polzin, 1999). The implication is that any explanation of bargaining
and firm asset sharing and labor stakeholding must also include an explanation for unions.
The empirical divergence may result from the focus on the interests and conditions of the
firm and employees and less attention on the features and mechanisms of property rights and
economic institutions. Across disciplinary areas, property rights and economic institutions have
become more significant in explaining economic decision-making and include rational-material,
distributive, and cultural-cognitive perspectives. Under rational-material conceptions, scholars
have posited that property rights and economic institutions determine who has residual control
and that these institutions efficiently support investments, such as in human capital (Grossman
and Hart, 1986; Hart and Moore, 1988, 1990; Rajan and Zingales, 1998).
Yet scholars under
distributional conceptions suggest that property rights and economic institutions distribute power
and determine outcomes by affecting the availability of authorized property forms, creating
market and corporate actors and establishing possible economic relationships (Fligstein and
Fernandez, 1988; Fligstein, 1996; Carruthers and Ariovich, 2004; and Campbell and Lindbergh,
1990; Marx, 2011).
Moreover, cultural-cognitive approaches to economic institutions may
question both of these previous accounts by arguing that social processes involving nonnative
and cultural approaches to shaping rationality may be as decisive in the adoption of property
rights and economic institutions. The present research
examines these hypotheses
and
emphasizes the importance of distributive conceptions of economic institutions and property
rights on the adoption of firm asset-sharing and employee stakeholder arrangements.
4 1Asset Sharing and Stakeholder Arrangements
The present research focuses on the 1985 reversal of Michigan's enforcement of
covenants not to compete. This reversal in enforcement allows us to examine an exogenous
change in economic institutions without an actual change in human capital. The finn asset
sharing and employee stakeholder arrangements of interest, ESOPs, were created by the U.S.
Congress in the 1970s as a savings and retirement supplement due to questions about the
coverage and sufficiency of Social Security entitlements (Freeman and Knoll, 2008). Over $928
billion in assets are held through broad-based stock ownership, covering twelve million persons
and over 10,800 companies (NCEO, 2011). ESOPs are important property rights and economic
institution for sharing firm assets with employees and thus make a good proxy for understanding
the distributive impact of changes in property rights and economic institutions.
To explore these hypotheses, I utilize both a quasi-experimental model which involves a
Difference-in-differences with synthetic controls as well as a count model with panel design. In
the quasi-experimental setting involving the random change in Michigan's covenants not to
compete law, I find that the enforcement of covenants not to compete decrease the adoption of
ESOP plans. On the other hand in a broader count model with panel design, I find that
unionization is not associated with an increase in the adoption of ESOP plans and, thus,
contravenes classical theoretical expectations on bargaining power.
I argue that the findings
related to covenants differ from rational expectations, which would expect a general increase in
compensation due to the risk of a significant wage loss from the covenants. I argue that both the
covenants and union bargaining scenarios may be more adequately explained by distributive
conceptions of economic institutions and property rights which serve to segment, allocate, and
distribute powers differently and, thus, result in diverging outcomes.
In addition, I also examine
cognitive-cultural approaches to property rights and economic institutions, which show that the
51 Asset Sharing and Stakeholder Arrangements
change in laws affected the adoption of these arrangements in regions around the state.
The key contribution of the present research is to provide an empirical setting to
document the impact of both bargaining power and distributive conceptions of property rights
and economic institutions on labor stakeholder arrangements and firm asset-sharing.
Second,
this research also seeks to deepen theoretical perspectives on the mechanisms related to
distributive conceptions; in this case, how the distribution of powers affects distributive
outcomes. Following the sociological literature, I argue that how economic agents and property
objects are constituted by economic institutions may influence how distributions of bargaining
powers align with the interests of institutionally constituted actors and whether they affects
bargaining
outcomes.
These distributive mechanisms of property rights and economic
institutions are particularly important because they suggest that bargaining power itself is not a
'clean'
indicator, but may be shaped by the configuration of property rights and economic
institutions.
Generally, the argument proceeds as follows. Section 2 examines the literature on human
capital and bargaining power in the adoption of employee stakeholder arrangements. It highlights
key theories related to property rights and economic institutions and proposes that rational,
distributive, and cultural-cognitive perspectives are important for the present discussion. Section
3 develops my estimation approach based upon the quasi-experimental setting of Michigan's
change in law as well as a broader count and panel model. Sections 4, 5, and 6 detail the
findings, conclusions, and limitations of the present research.
6 Asset Sharing and Stakeholder Arrangements
0.2 Theory
Labor
Stakeholder
Arrangements,
Theories
of
Human
Capital Investments
and
Bargaining Power
Preserving and guarding investments in human capital have emerged as key motivations
for firm asset sharing and adopting employee stakeholder arrangements, such as broad-based
employee ownership. The argument is that firms and employees need a way to protect both
employee and employer investments in unique, firm-specific human capital (Blair, 1999).
Theoretically grounded in the economic literature, these arguments rely on the distinction
between general versus firm-specific skills (Becker, 1962, 1975). While general skills are those
which contribute to a worker's general human capital, firm-specific skills are those skills which
increase a worker's productivity with only the current employers. Becker (1962, 1975) has
previously argued that there would be no firm-sponsored investments in general skills training
because the worker could egress with the skills to another firm. Yet, empirical evidence suggests
that firms do make general investments in workers such that they may have a real interest in
protecting these investments (Autor, 2001; Acemoglu and Pischke, 1997, 1998; Pendleton and
Robinson, 2011). The key problem with specific skills is parallel to that of general skills. If the
worker makes investments in firm-specific skills, the result is an increase in firm profits. As
such, the firm wants the worker to make these investments; however, the worker, knowing that
he is not able to gain in the increases in profit, may not commit to make these investments.
Therefore, the general issue for human capital investments is how to enable both firms and
individuals to make and protect their mutual investments-whether general or specific-in
human capital.
Scholars argue that the ownership or property rights in the finn are non-contractual
7 Asset Sharing and Stakeholder Arrangements
arrangements that can enable and protect these mutual investments in human capital (Blair, 1995;
Robinson and Zhang, 2005). The evidence suggests broadly that employee ownership is
associated positively with human capital. Robinson and Zhang (2005) find that workplaces with
employee stock ownership report higher levels of human capital than firms without employee
stock ownership, noting that "as human capital elements in a firm become more important,
employee membership of ESO [employee stock ownership] increases." They find that employee
stock ownership arrangements are more likely to be found in workplaces that depend on
employees to make valuable investments in human capital (Robinson and Zhang, 2005).
Moreover, Pendleton and Robinson (2011) find that employee share plans are associated with
higher levels of training activity and high levels of participation in these plans. Ultimately, these
models suggest that a positive relationship should exist between human capital intensiveness and
the adoption of broad-based stock ownership.
While the literature on broad-based stock ownership has utilized the literature on human
capital investments as a reason for the adoption of these arrangements, related literature has
highlighted the mechanism of bargaining power, which is particularly relevant in the contexts of
incomplete contracts. Stylistically, in the complete contract setting, wage bargaining in the first
stage determines the wage for the whole duration of the relationship, and employers can
unilaterally determine investment in human capital (Cahuc and Zylberberg, 2004). Yet, Grout
(1984) argues that in incomplete contracts settings, employers make an investment in the first
stage, and then negotiate wages and benefits afterwards. Here, bargaining power determines the
share of the surplus caused by investments in firm-specific human capital that goes to the
worker. Similarly, Aoki (1980, 1984) and others have modeled the firm as a profit-sharing entity
between shareholders and employees, departing from neoclassical models of the firm (Moretto
8 1Asset Sharing and Stakeholder Arrangements
and Rossini, 1995, 1996; Jou and Lee, 2004).
Similar to human capital investment theories,
these firm-based models suggest that employees "embody skills and knowledge more or less
specific to the firm as a result of quasi-permanent association with it" (Aoki, 1980). Aoki (1980)
notes that the quasi-permanent associations between employer and employee produce unique
economic gains as organizational rent. Employees can exert implicit and explicit bargaining
power to split organizational rents and obtain premium earnings above wages (Aoki, 1980, 1984;
Moretto and Rossini, 1995, 1996). These premium earnings with firm-asset sharing -
which we see as coterminous
are determined through a game between employees and employer,
arbitrated by a manager or collective bargaining (Aoki, 1980, 1984). As a classical heuristic,
bargaining outcomes thus hinge upon the bargaining power of the stockholders and of the
employees.
As extensions of bargaining power theories, strategic management and employee
stakeholder theorists provide a parallel account of how bargaining power affect changes in firm
governance. Coff (1999) has examined factors that increase the bargaining power of employees
including capabilities for unified action, access to information, replacement costs and costs of
exiting to stakeholders. Other scholars have noted how the shift to knowledge intensive work
increases the balance of power to workers and thus the need for firm asset sharing arrangements
(Rousseau and Shperling, 2003; Roberts and Van Steen, 2000). Kochan and Rubenstein (2000)
note the insufficiency of previous stakeholder literature that suggests that managerial values and
leadership style are key predictors of the emergence of employee stakeholder institutions in
firms. They contend that that employee stakeholder arrangements in firms would emerge when
potential stakeholders: (1)
supply critical resources or assets to enterprise, (2) when value of
assets are affected by fate of enterprise, thus, creating property right claims, and (3) when
9 1Asset Sharing and Stakeholder Arrangements
potential stakeholder have adequate power to challenge privileged power (Kochan and
Rubenstein, 2000). These bargaining power theorists similarly suggest that bargaining power
may be as important a mechanism as protecting specific investments in human capital.
Firm Asset-Sharing, Labor Stakeholder Arrangements
and the Role of Economic
Institutions and Property Rights
Generally, the emphasis of previous theories on employee stakeholder arrangements and
firm asset sharing has been on what features of firms and employees are necessary to impel their
use.
In one case, the interests of firms and employees were important, that is, to protect
investments in human capital. In the other case, the role of bargaining leverage in affecting the
outcomes between the two was key.
Yet, we believe these stories
are insufficient because
inadequate emphasis has been given to the subtleties of economic institutions and property rights
in the adoption of employee ownership arrangements. There is a broad consensus that economic
institutions affect economic performance. Yet, economic institutions and property rights come in
varying forms across national and subnational contexts (Kennedy, 2011). Even the same
economic institution -
the corporation, the cooperative, the union, employee institutions -
have different legal and institutional forms (Bagchi, 2007; Kennedy, 2011).
can
Moreover, other
relevant laws and economic institutions may curtail or complement the use of particular property
rights institutions. From the previous institution-less theories, the assumption would be that the
appropriate property rights would always be available, based upon firm-worker interests and
conditions that impel these two mechanisms. While previous accounts have explicitly
acknowledged the role of ownership, they underplay the forms and mechanisms of how
economic institutions and property rights interact with economic decision-making processes.
Generally, scholars across disciplinary perspectives have been keen to the role of
10 1Asset Sharing and Stakeholder Arrangements
economic institutions and property rights in shaping economic processes and decisions.
Traditionally, scholars examined economic institutions as having a regulatory effect on markets
that were in some sense naturally occurring; these expositions saw economic institutions as
attending to market failures in the spirit of Breyer (1982) where justification for regulations were
addressing monopoly power, excess profits, spillovers, inadequate information, or excess
competition.
Moreover, many of these approaches emphasize how these regulatory regimes
arise, either through interest-driven action, efficiency, regulatory capture, or other mechanisms.
Later expositions of economic institutions and property rights have moved beyond an emphasis
on economic institutions as regulative of market interactions to seeing economic institutions as
constituting markets themselves often because of the political, social, and institutional processes
involved in creating the economy (Edelman and Suchman, 1997; Dobbin and Dowd, 1997).
Constitutive approaches extend beyond regulative approaches, by seeing economic institutions as
"taken-for-granted labels, categories, and 'default rules'... [which] determine what types of
organizations come into existence and what type of organizational activity gain formal
recognition" and "establishes which organizations can act as corporate 'persons' and which can
act only as collections of separable interests." (Edelman and Suchman, 1997) With this focus on
how economic institutions and property rights constitute markets, three perspectives have
emerged in how economic institutions and property rights shape the economy. One perspective,
parallel to the incomplete contracts perspectives presented above, views economic institutions
from rational-material perspectives where transaction costs and markets are key. The culturalcognitive perspectives emphasize economic institutions and property rights as social processes
emphasizing normative, cultural, and isomorphic concerns.
A third perspective, paralleling
bargaining power perspectives, emphasizes economic institutions and property rights as
111 Asset Sharing and Stakeholder Arrangements
distributive institutions. These understandings about the institutional environment-particularly
about economic institutions and property rights as constitutive and encompassing the three
perspectives-have not been fully utilized in understanding the adoption of firm asset sharing
and stakeholder arrangements (Kaarsemaker et al, 2009). We argue that an improved explanation
for the adoption of employee stakeholder arrangements and firm asset sharing should understand
the role of economic institutions and property rights in affecting these economic processes and
decision-making.
Rational-Material Perspectives
Rational-material perspectives on economic institutions and property rights focus on their
role in responding to the operation of markets and its related costs.
Edelman and Suchman
(1997) argue that the transaction cost literature anchors a rational-material perspective to the
constitutive dimension of economic institutions by addressing "the role of legal categories,
capacities, and 'default rules' in organizational life."
In transaction cost theories, extensive costs arise in the market from learning and
haggling over terms of trade, which can be significant if the transaction is long term (Hatt,
1989). Moreover, the complexities of specifying the contracts in detail will result in contractual
incompleteness and thus leave room for ex-post bargaining. Williamson (1971) argues that
transaction costs are particularly important where actors make relationship specific investments.
These investments as "the fundamental transformation"
are important because after the
investments are made, parties are locked in with each other and, thus, markets may not provide
adequate information about opportunity costs.
One general way to deal with contractual incompleteness and reduce costs is by allowing
one party to have authority and suppressing the price mechanism (Hart, 1989). Grossman and
12 1Asset Sharing and Stakeholder Arrangements
Hart (1986) argue that ownership is the key to firm integration because a firm is composed of
assets, and ownership entails the residual rights of control of an asset.
Importantly, these
ownership rights influence ex-post distribution and thus affect ex-ante investments (Grossman
and Hart, 1986).1
The general idea is that she who holds ownership will know ex-post
distributions and thus be able to make the appropriate ex-ante investments.
Ownership and its incentives maximize investment in human capital (Zingales, 1998;
Rajan and Zingales, 1998).
Effectively, the property rights approach to the firm parallels
arguments previously presented about protecting investments in human capital. The underlying
assumption is one of incomplete contracts where ownership serves to resolve ex-post difficulties
and encourage ex- ante investments. The broad hypothesis from this literature is that economic
institutions and property rights would continue to operate to protect investments in human
capital.
Cultural-Cognitive Perspectives
A second set of approaches to property rights and economic institutions, broadly
embraced as sociological institutionalism, view institutions as cultural-cognitive manifestations
(Fligstein and Choo, 2005). The core of institutional theory in sociology is that structures are
symbolic phenomena where rationality is socially constructed and, therefore, less focused on the
efficiency aspects of property rights and economic institutions (Dobbin, Sutton, Meyer, and
Scott, 1993).
These neo-institutional approaches envision open systems where cultural and
Hart and Moore (1990) examine the effects of ownership rights on employees, rather than management.
They argue that workers' efforts are often asset-specific such that actions taken today toward productivity only pay
off if they have access to the asset tomorrow. Contract incompleteness results in the worker's current action being
dependent on marketability.
13 1Asset Sharing and Stakeholder Arrangements
nonnative elements from the institutional environment influence the internal structuring of
organizations (Sutton et. al, 1994).
Early accounts of neo-institutional theory emphasize how organizations conforn
to
legitimate models of highly institutionalized environments (Meyer and Rowan, 1977; DiMaggio
and Powell,
1983;
Sutton and Dobbin,
1996).
Under the constitutive view of neo-
institutionalism, Edelman and Suchman (1997) argue that at least three views have emerged:
normative, cognitive, and coercive accounts of institutions.
Guided by early neoinstitutional
accounts, normative perspectives see institutions as providing normative values and ethics which
guide individual actions (Selznick, 1996; Edelman and Suchman, 1997).
Cognitive accounts
see institutions as effecting action because they are more natural, legitimate and appropriate.
Finally, coercive involve pressures through politics and law; however, these pressures are hard to
distinguish from more rational or material perspectives (Edelman and Suchman, 1997).
A number of neo-institutional scholars have empirically examined the role of economic
institutions in structuring the employment relationship and the firm.
Some scholars have
examined how law and public policy interacts with organizations and employment relations
focused on contexts where there was some type of legal threat or uncertainty that needed
resolution. Edelman (1990) argues that threats posed by the legal environment, through direct
and indirect effects, caused organizations to implement due process procedures that looked like
legal models of governance
Dobbin, Sutton, Meyer, and Scott (1993) examine how public
policy affects the construction of internal labor markets. They argue that equal employment
opportunity law, personnel managers, and courts jointly shaped internal labor markets as
legitimate ways to prevent discrimination and enable effective allocation of human capital
(Dobbin et. al, 1993).
They propose that while policy and law seldom dictate individual
14 1Asset Sharing and Stakeholder Arrangements
behaviors, they can lead to modes of organizational compliance (Dobbin et al, 1993).
Other
scholars argue that a shift in normative environment can shift how law is enacted. Edelman and
Suchman (1997) examine how the civil rights movement and mandates of the 1960s expanded
due process protections as organizations
sought to conform to the changed normative
environment. Others demonstrate how institutionalization can not only improve individual rights
in the economy, but also reduce individual rights in the economy.
For example, Sutton and
Dobbin (1996) demonstrate the role of institutional processes in the diffusion of grievance
procedures and employment at will laws, respectively rights-enhancing and rights-negating
systems.
The general hypothesis from sociological institutionalism is that economic institutions
and property rights operate through more cultural-cognitive mechanisms and social processes to
affect economic outcomes. Particularly, a shift toward a rights enhancing or constraining law
may encourage the adoption if the law, regardless of their effects.
Empirically, these effects
would be increasing over time. Moreover, we may see these effects in places where the law does
not coercively reach, but may normatively influence.
Distributive Conceptions
If previous claims that bargaining power affects bargaining outcomes are true, it may
suggest a conception of economic institutions and property rights as more distributive. Scholars
across disciplines have argued for the importance of distributional conflict, noting that it should
be the centerpiece of institutional analysis (Knight, 1997). They argue that property rights and
the organization of economic production are largely distributional because they are a product of
social conflict (Knight, 1997). In some approaches, the distributional effects of institutions are
seen as a key mechanism to support feedback effects, which facilitate the organization and
15 1Asset Sharing and Stakeholder Arrangements
empowerment of certain groups at the expense of other groups (Thelen, 1999). While distributive
conceptions of property rights and economic institutions originate from different disciplinary
locations and have different explanatory purposes, they contain a few core assumptions about the
nature of property rights and economic institutions as central to distributive outcomes.
These
common assumptions have important implications for theories of the firm as well as stakeholder
and firm asset sharing theories. These common assumptions of distributive conceptions of
property rights and economic institutions include: (1)
property rights are unequally distributed,
(2) property rights are pre-configured with state legitimization, and (3) property rights allocate
power.
Most distributive conceptions of property rights and economic institutions assume that
property and economic institutions are pre-distributed. This is reflected in an extensive literature
related to the role of unequal ownership in societal inequality. Early classical scholars saw the
ownership of property and the means of production as key distinguishing and grouping
mechanisms between individuals (Webber, 1978; Marx, .n.d). A broad set of studies see wealth
inequality, through property ownership, as having important micro-level effects and broader
macro-level effects. Sociologist and political scientists examine the inequality of property
ownership on political power, mobility, life conditions, and conflict (Spilerman, 2000; Sorenson,
2005; Lupu and Pontusson, 2011; Kentworthy and Pontusson; 2005). Scholars have also noted
how previous wealth, often linked with other group characteristic, affect future wealth and
inequality through inheritances, return on assets, and savings patterns (Spilerman, 2000;
Gittleman and Wolff, 2004; Shapiro and Oliver; 2006).
Similarly, classical economic theory
views individual behavior as affected by the individual's level of starting wealth, with many
studies analyzing wealth limitations under the concepts of credit and liquidity constraints.
16 1Asset Sharing and Stakeholder Arrangements
Liquidity constraints assume that there are imperfections in capital markets which limit the
activities of those with low wealth. A number of studies agree that liquidity constraints bind the
behavior
of
those
entrepreneurialism
with
low
wealth,
upset
intertemporal
consumption,
and entrepreneurial survival (Evans and Jovanovic,
Leighton, 1989; Hotlz-Eakin, Joulfaian, Rosen, 1993)
and
affect
1989; Evans and
2
A general conclusion of the literature on the pre-distribution of property rights suggests
that various structural limitations are imposed on economic processes and decision-making.
First, pre-distributed wealth means that individuals may lack the means to enter into efficient
bargaining relationships, have fewer options for creating productive
and entrepreneurial
relationships, and may deter employees from exiting troublesome employment relationships
because of the role of wealth in enabling consumption during non-working periods
These
assumptions contrast with assumptions of perfect credit markets where previous property right
distributions would be unimportant in economic decision-making. They also contrast with
assumptions of efficient bargaining.
Second, most distributive theories of property rights and economic institutions highlight
that previous conflicts shape them into specific forms and configurations.
Most interest-based
political economy theories of state and economy emphasize the role of particular actors and
groupings of actors in affecting the structuring of state and economic institutions (Swenson,
1991; Esping-Anders, 1990; Swenson, 1990; Estevez, Iverson, and Soskice, 2001; Hacker and
Pierson, 2010). Political scientists under a distributive conception argue social conflict is key in
producing the distributional effects of property rights and the organization of economic
production (Knight, 2004). Sociologists note how conflict shapes economic institutions and labor
2 Yet, other economists argue that the costs of entering entrepreneurship and self-employment are low and thus not
real barriers (Hurst and Lusardi, 2004).
17 Asset Sharing and Stakeholder Arrangements
market institutions (Sorenson, 2005; Wright, 2000; Fligstein and Fernandez 1988; Fligstein,
1996).
Other scholars argue that previous distributional outcomes influence de facto political
power and thus the ability to structure future economic institutions (Acemoglu, Johnson, and
Robinson, 2005). A key difference in these perspectives often involves the mechanisms of the
social conflict and the nature of institutions, but all agree that economic institutions are shaped
and configured by this conflict.
Implicit in this pre-configuration of property rights is the role of the state in legitimizing
these configurations of property rights and economic institutions. Economic relations are
traditionally conceived of as dyadic relationships between actors in the contractual view;
however, because economic institutions involve shaping and enforcing power of the state, these
relationships involve more triadic relations of the state (Carruthers and Ariovich, 2004). In fact,
sociologists see property rights as giving the state, even a fragmented U.S. state, an unrecognized
source of strength within the economy (Campbell and Lindbergh, 1990). Due to state
legitimization and authorization, property rights are pre-configured so that they are somewhat
resistant to change, are enforced by the agents of the state, and implicate the state in maintaining
or generating the inequality effects of property rights and economic institutions (Campbell and
Lindbergh, 1990; Sorenson, 2005).
The implication of this pre-configuration view is that
property rights and economic institutions, once established, are hard to change before, during, or
after bargaining. Therefore, the contracting and bargaining space are as likely to be occupied by
pre-written, structured contracts based upon the property rights order.
A third key assumption of distributional conceptions is that property rights and economic
institutions are tightly coupled to processes of power (Campbell and Lindbergh, 1990; Carruthers
and Ariovich, 2004; Acemoglu, et. al 2005).
18 1Asset Sharing and Stakeholder Arrangements
Scholars have traditionally viewed power over
things as allocations of power over people because owners of productive property can prevent
others from using it (Marx, n.d; Weber, 1978; Campbell and Lindbergh, 1990; Carruthers and
Ariovich, 2004). The links between economic institutions and power have long been explored
within particular economic institutions. Labor control theorists argue that organizational rules
serve "to contain and appropriate labor power" (Clegg, 1981; Sutton and Dobbin, 1996).
Sociological rent theorists argue that a change in laws and conceptions of the firm
as an
economic institution shift power between economic actors, for example within the financial
sector which enabled them to accumulate a larger share of national income (Sorenson, 2005;
Tomaskovic-Devey, and Lin, 2011).
Others examine how state licensing, credentialization,
certification, and unionization serve to increase occupational power through non-market
mechanisms of closure whereby workers are able to limit the supply of other workers (Weeden,
2002). More recently, political economists have argued that distributive outcomes arising from
economic institutions shape political processes and thereafter determine the future structuring of
political institutions and the shape of economic institutions (Acemoglu, et al, 2005).
Placed against this literature, the present inquiry highlights two constitutive processes of
property rights and economic institutions and how they distribute powers: (1)
property objects and (2) by creating economic actors and agents.
by defining
Economic institutions allocate,
segment, and distribute powers by defining property and economic objects.
Carruthers and
Arivioch (2004) highlight the basic dimensions of property rights as (i) what can be owned, (ii)
who may own, (iii) what can be done with property, and (iv) who can enforce rules related to
property. Each of these dimensions represents formal allocations of powers into and over objects
and into social relationships.
The relevant economic objects are varied as societies permit
different things to be owned based on varying cultural and moral conceptions (Carruthers and
19 1Asset Sharing and Stakeholder Arrangements
Arivioch, 2004).
Importantly, these objects do not have to be physical as the modem shift in
property rights has been from only traditional ownership (land and physical objects) to intangible
objects (bonds, patents, trademarks) (Carruthers and Ariovich, 2004).
However, these dimensions emphasize that property rights and economic institutions
determine which actors can do what vis-a'-vis other actors with these economic objects. Property
rules determine who can establish a property object, how it can be used, whether it can be passed
to others, and who can enforce (Carruthers and Ariovich, 2004). Additionally, property rights
may constrain or enable activities by excluding powers or arrangements that would have been
contractually preferred or, alternately, creating more extensive powers.
An important aspect of how property rights allocate power is not unitary, but may be
segmented around various dimensions of a property object and toward specific purposes. As
described earlier, the state often is key in enforcement of property relationships, but also often
shapes standards and precedent related to how powers can be used 3 . Moreover, legal articulations
around the use of property may restrict or expand what may be done with property in relationship
to other individuals (Carruthers and Ariovich, 2004). For example, Citizens United v. Federal
Election Commission, 558 US 310, (2010) now permits corporations to use their treasury funds
for direct political advocacy, thus expanding corporate power in U.S. elections.
In addition to creating property objects, scholars have also argued that economic
institutions are important in how they allocate powers to create new economic actors and agents
(Fligstein and Fernandez, 1988; Fligstein, 1996). Scholars highlight the importance of economic
3 Functionally, the creation of property or economic objects may serve multiple purposes. Without this state
allocation of powers to property objects, certain arrangements may be too costly to create. Secondly, they may serve
to provide categories and symbols for economic sense-making. The availability of particular property rights may
provide the identification, consistency and certainty that traditional contracting may not address. Moreover, they
may determine status of individuals in relationship to each other. Yet, the present argument emphasizes that
property and economic objects represent particular allocations of legal powers granted to certain actors over an
object or context in relationship to other actors, all of which is provided some level of enforcement by the state.
20 1Asset Sharing and Stakeholder Arrangements
institutions in defining and establishing economic actors and agents, from corporations to unions,
which influence the shape of labor market bargaining (Fligstein and Fernandez, 1988; Fligstein,
1996). Others note the role of economic institutions in enabling large corporations and private
organizations to dominate economic governance (Campbell and Lindbergh, 2004).
Legal
scholars have argued that corporate forms are more than mere aggregation of individuals, but
that corporate forms, such as corporations, contain unique legal allocation of powers, which may
include legal personhood, perpetual status, and for private entities limited liability and other
unique entitlements (Blair, 1999; Hansmann and Kraakman, 2000). Similar to the creation of
new objects, the creation of new economic actors determine the powers of actors through similar
dimensions as property objects, but often with more expansive powers.
Importantly, these state and institutional allocations of powers to create economic agents
result in unique configurations of interests (Swedberg, 2003).
These interests arise from how
allocations of legal power configure the features and processes of these economic actors. One
clear example is the allocation of powers related to governance. A key difference between the
corporation and the union is the allocation of governance power.
A union may have legal
powers allocated on the basis of membership through a one member-one vote basis.
Corporations are governed by quasi-ownership rights based on the level of equity investment,
even though it may separate formal ownership and control. These differences in allocations of
power effect governance and structuring and, thus, produce systematic differences in interests.
One premise that results from the convergences of these two processes of creating
property objects and economic agents is that the creation of new property objects allocates
powers differently between actors. The assumption is that a property object may be better suited
to one agent's interests more than another agent. Similarly, it may be categorically better suited
211 Asset Sharing and Stakeholder Arrangements
to a set of actors' interest versus another set of actors, for example firms versus individuals or
unions versus firms. There may be certain complementarities between the legal powers and
purposes of the economic actor and the property object.
However, it is not necessary that a property right be utilized for its power to be allocated,
but simply that it exists. Some configurations of powers and property rights are extremely
difficult or impossible to contract without the formal existence of the property right or economic
institution. Without corporate law and the state-provided powers of limited liability, perpetual
existence, and other features, corporations would be less feasible. And, because of their
existence, actors with congruent interests are likely to utilize these property objects. Under
neoclassical
assumptions,
economic
relationships
would
involve
primarily
individuals
contracting with each other. However, the definition of economic objects and economic agent
shifts the alternatives available for realizing specific interests and purposes as well as the
arrangements of powers of one set of actors in relationship to others. For example, the wealthy
can finance large corporations which they govern financially rather than use simple or complex
contracts to advance their economic purposes. This legal existence assumption aligns with
power-dependency conceptions which typically depict power as inversely proportionate to the
number of alternatives (Emerson, 1962). The argument, therefore, is that the mere existence or
creation of a property right begins the process of reallocating and distributing powers among
economic agents.
Within the present argument about stakeholder arrangements and firm asset-sharing, the
availability of appropriate stakeholder and firm asset-sharing property objects would be
positively associated with the increased adoption of these mechanisms. The idea is that where
appropriate stakeholder property rights exist, there will be an increase in the types of powers that
22 1Asset Sharing and Stakeholder Arrangements
support firm asset sharing and employee stakeholding. The correlate is that the lack of a set of
property rights or the availability of substitute economic institutions would reduce the adoption
of these property rights forms.
Effectively, the mere availability of substitute economic
institutions would reallocate legal powers from firm asset sharing and employee stakeholding.
Therefore stakeholder-reducing or substituting property rights are likely to decrease bargaining
power related to the adoption of stakeholder arrangements.
23 Asset Sharing and Stakeholder Arrangements
0.3 Research Context
Description of Data Set
To test how distributive conceptions may shape bargaining power in stakeholder
arrangements and firm-asset sharing, I create a dataset of employee stock ownership plan
adoptions for 50 U.S. states for the period 1981-1990. This results in a data set with 500 stateyears and 12,980 firm-years. This time period is utilized because it captures the period during
which the key legal changes occurred, but also precedes a significant change in the laws
regarding broad-based stock ownership. The dependent variable, the adoption of ESOPs, was
obtained from the National Center for Employee Ownership, which collects data from U.S.
Department of Labor's (DOL) Employee Benefits Security Administration. We supplement this
dataset with information on state economic conditions from the Current Population Survey, the
Bureau of Economic Analysis of the U.S. Department of Conunerce, and the Longitudinal
Business Database of the U.S. Census. In addition, I reorganize a streamlined dataset of over
882,886 observations in the IPUMS American Community Survey to create state by time proxies
for highly educated and highly skilled indicators for educational and occupational groups.
In the following sections, I highlight the key variables of interest including broad-based
stock ownership, covenants not to compete, and unionization measures, emphasizing predictions
from a distributive conception of economic institutions and private property, and particularly
where they may contrast with classical bargaining power predictions. In addition, I also briefly
highlight key covariates and controls utilized.
24 1Asset Sharing and Stakeholder Arrangements
Employee Stakeholder Arrangements and Firm Asset Sharing - Dependent Variable
I utilize the adoption of ESOPs as the central dependent variable for employee
stakeholder arrangements and firm-asset sharing. ESOPs emerged under Congressional concerns
over the viability of the Social Security System (Freeman and Knoll, 2008). Originating under
the leadership of San Francisco Attorney and investment-banker Louis Kelso and U.S. Senator
Russell Long of Louisiana, these plans aimed to broaden the ownership of stock among
employees, increase performance of the companies, and spur the productivity of the employees
(General Accounting Office, 1986). As a supplement to Social Security, ESOP legislation was
part of the Employment Retirement Income Security Act (ERISA) of 1974 (Freeman and Knoll,
2008).
The act permits companies to establish ESOP trusts that hold the stock of a company on
behalf of the plan's participants. The company is able to appoint a fiduciary who manages the
plan on behalf of its participants. Additions to the Act have included
a number of
nondiscrimination requirements to ensure that most employees of a firm have an opportunity to
participate in plans, that the plans operate in their interests, and that a significant time period is
not required for vesting.
ESOP legislation also includes a number of benefits and tax incentives. Leveraged
ESOPs can uniquely borrow money to buy shares. The company's contributions of stock to
ESOPs are tax deductible, direct cash contributions are deductible, and any repayment of a loan
taken out by the ESOP to buy the company's shares are tax-deductible and done in pre-tax
dollars. Moreover,
dividends are tax-deductible. There
also are
additional possibilities,
depending on the corporate form, for other types of tax deferrals and having the ESOP shares not
being taxed at the federal level. Employees pay no tax on the contributions to the ESOP, only the
25 Asset Sharing and Stakeholder Arrangements
distribution of their accounts, and then at potentially favorable rates. The significant tax benefits
of the ESOP make it both a way to provide stock ownership to employees, but also a corporate
finance option. The plans can be used to provide working capital, carry out a recapitalization,
buy out a larger shareholder or finance investments.
As mentioned previously, for this research, I utilize the adoption of ESOPs within all
U.S. states for the period 1981-1990. Firms are required annually to file form 5500 with the IRS,
which includes
information on plan initiation, location, participation,
distributions, and
contributions. As part of its ERISA responsibilities, the DOL's Employee Benefits Security
Administration, the Internal Revenue Service, and the Pension Benefit Guaranty jointly
developed this form and collect the data, which is generally maintained for seven years. The
National Center for Employee Ownership has annually collected this data from the U.S. DOL to
create a combined dataset of all firms with ESOPs. The present dataset creates a balanced panel
of 50 states with the numbers of these plans initiated for each state from 1981 to 1990, which
generates 500 state-year observations. Figure 1 and 2 provide broad perspective on the
distribution of these plans and their adoption during the 1981-1990 period. During this period,
firms within the U.S. adopted a total of 1,298 trusts. As shown in Table 1, the average state has
2.596 trusts adopted with a standard deviation of 2.989. The maximum number of trusts adopted
during this period is 15. The average stock held by each plan is 8 million shares. The average
number of people in each plan was 10,401 persons. Figure 1 demonstrates that the adoption of
ESOPs has a distribution similar to count variables. Particularly relevant for this research is
Figure 3, which demonstrates the change in the adoption of ESOPs during the 1985-1987 period.
A key issue in this project is whether the change in the enforcement of covenants not to compete
caused the change in adoption of ESOP trusts.
26 Asset Sharing and Stakeholder Arrangements
Covenants Not to Compete
The present research examines the 1985 change in Michigan law, pennitting the public
enforcement
of covenants
not to compete.
Covenants
not to compete
("non-compete
agreements") are agreements, as a condition of employment, that restrict the activities in which
an employee can engage after his/her employment ends. These agreements aim to protect the
interests of the employer. They are governed by state contract law and thus vary across the
country. The basic legal conflict in these agreements emerges within the opposing goals of
freedom to contract and contractual restraints on trade (Pynnonen, 1994).
Many states have allowed their courts to rule on non-compete agreements, but Michigan
chose to leave these decisions to legislators.
This decision resulted in the Michigan statute
having not changed from 1905, when these covenants were ruled void, until 1985 (Pynnonen,
1994; Marx, Strumsky, and Fleming, 2009). In 1985, the Michigan legislature repealed the
previous statute and permitted enforcement of covenants not to compete under the Michigan
Antitrust Reform Act.
A number of researchers have examined the role of non-compete agreements on
employee mobility, human capital development, and adoption of broad-based stock ownership.
Marx, Strumsky, and Fleming (2010) argue that non-compete agreements affect employees by
decreasing their mobility. They also find that with enforcement of non-compete agreements,
workers who have developed firm-specific human capital are 15.4% less likely to change
employers (Marx et al, 2010). Other researchers have found that in enforcement regions, when
individuals with greater human and social capital leave their companies, they are also likely to
leave the region (Marx, et al, 2010).
Marx (2011) finds that non-compete agreements shape
labor markets through "career detours" where individuals change industry for fear of being
27 Asset Sharing and Stakeholder Arrangements
prosecuted and "seek shelter" where individuals work for large entities to protect against legal
In terms of significance, Marx (2011) argues that nearly half of all technical
claims.
professionals are subject to non-compete agreements
by employers, thus suggesting its
importance.
Garmaise (2011) develops an index of non-compete agreement variation by state and
finds that the stronger enforcement of non-compete agreements ensures executive stability, shifts
executive compensation toward the use of salary, and reduces capital expenditures per employee.
Relative to our interest, Garmaise (2011) did not find that the use of stock options for executives
varied with noncompetition enforcement.
Kedia and Rajgopal (2009) examine the impact of
location on the use of broad-based option grants. They find location matters through state use of
non-compete agreements, firm stock price movement, and the highly educated workforce. While
Kedia and Rajgopal (2009) utilize Garmaise (2011) index on non-compete variation, they do not
investigate, theoretically, the distributive role of economic institutions as an explanation for the
impact of non-compete agreements. Moreover, our findings utilizing the Garmaise (2011) index
demonstrate that it does not provide any strong conclusion on the effect of covenants not to
compete.
Unions and Employee Stakeholder Arrangements
Unions figure strongly into the bargaining power of workers. Not only do they bargain
directly for increasing wage and compensation benefits for the less educated, younger, and junior
workers, but they also impact non-union firms who seek to avoid unionization by implementing
higher wages and better working conditions (Freeman and Medoff, 1984; Farber, 2005). Scholars
argue that the union decline explains a significant part of the decline in wage growth across
OECD countries as well as rising inequality (Western and Healy, 1999; Western and Rosenfeld,
28 1Asset Sharing and Stakeholder Arrangements
2011). The U.S. has the lowest ranking among OECD countries with 16% of U.S. workers being
in unions and 18% covered by collective bargaining contracts.
Scholars note that labor unions have maintained a mixed relationship with employee
stakeholder arrangements, such as ESOPs, because these plans have been more associated with
firm interests than worker interests.
Early views found that firms wanted to use employee
ownership to reduce unionization efforts; in welfarist versions of capitalism in the 1920s,
employee stock ownership was used as a way to block the unionization of workers (McCarthy et
al, 2011). Broad-based stock ownership may reinforce worker identity with the firm, which may
undermine industry-wide standards for labor (McCarthy et al, 2011).
Additionally, there are
concerns that firms may want try to unload unprofitable or failing firms onto workers (McElrath
and Rowan, 1992).
The union's pecuniary interests are also important. Unions prefer that their compensation
and benefits be incorporated in their collective bargaining contracts. Also, the National Labor
Relations Act requires that the compensation and accumulation of benefits feature of ESOPs
only be mandatory during collective bargaining. Unions often prefer financial compensation
programs that add to base pay and pension, rather than put pay and retirement at risk (McCarthy
et al, 2011). Moreover, the differential pay across facilities associated with employee stock
ownership may undermine the interests of equal pay for equal work that unions advocate
(McCarthy et al, 2011). Legally, firms are able to exclude union members from the plan (Bagchi,
2007).
Empirically, I utilize unionization rates within the state to control for the effects of unions
on broad-based stock ownership. This indicator allows us to control for state level effects of
29 Asset Sharing and Stakeholder Arrangements
unionization. However, it does not allow us to see the union specific mechanisms or systems
through which these effects may be occurring.
Additional Covariates
In addition, I utilize a number of covariates to capture the effect of varying aspects of the
state economy. I use a set of controls to capture key time variant factors for the state economy
including: the total population measured in individuals, the total state income measured in
dollars, net earnings per capita measured in dollars, wage and salary earnings in dollars, the
unemployment rate, and the per capita income. The population and total income measures allow
me to capture the effects of labor productivity. In order to capture the effect of human capital, I
utilize a few indicators. First, following Acemoglu and Autor (2012), I utilize the percentage of
above bachelor degree population. Second, I also utilize the percentage of professionals and
technical workers.
I also utilize a set of controls for the state of the financial markets. A key concern could
be other legal changes, which may affect the ability of firms to receive financing. Alternatively,
financial markets could have been experiencing significant fluctuations, which had some effect
on broad-based stock ownership adoption.
Following Morgan, Rime, and Strahan (2003), I
utilize the housing price index, which acts as a measure of collateral in the state. I utilize statebased indicators for financing because a significant number of states during this period were
more independent of national sources of finance (Morgan, Rime, and Strahan, 2003). In addition,
I utilize dividends and interests by state to capture similar financial effects. Jou and Lee (2004)
have extended Aoki's model from the monopoly context to the oligopoly context with
competitive pressures. Their conclusion is that a firm facing more competitors will grant lower
30 1Asset Sharing and Stakeholder Arrangements
shares of profits to employees (Jou and Lee, 2004). Therefore, I also utilize an indicator to
capture the population of businesses.
311 Asset Sharing and Stakeholder Arrangements
0.4 Research Model and Hypotheses
The use of an informal, conversational game model is helpful to bring together our
previous discussion of the literature to describe more specifically the effect of a change in law on
firm asset sharing and the adoption of employee stakeholder arrangements and to distinguish
between the three hypotheses.
Imagine the classical two-period model of bargaining. The first
period involves firms and workers contracting and making investments and the second period
involves production outcomes and allocation of benefits.
In order to fit the present theory, we add a period zero which involves the random choice
of legal regime. In time period zero, the legal regime is established with one regime representing
only ESOPs and another with ESOPs and covenants not to compete. In the first period, workers
and firms contract and make investments. At this stage, firms contract with wages plus some
share of the profits and the firm.
Firms share profits through the adoption of ESOP plans. In the
second period, production occurs and the firms and workers receive their share of the profits.
Workers receive the share of the profits and the firm according to their bargaining power.
With firm specific human capital, firms want to incentivize workers to make investments
in firm specific human capital. Yet, workers know that the benefits of these specific investments
will go only to the firm because they are not valuable in the market. Using backward induction,
the firms contract via ESOPs in the first period as firm asset sharing devices and as credible
commitments not to hold up the worker. In the second period, workers receive the allocations
through these ESOPs. With general investments in human capital, firms realize that any general
investments made in human capital will walk away from the firm in the second period. This is
because general investments in human capital are reflected in an increase in the worker's market
32 Asset Sharing and Stakeholder Arrangements
wages.
Typically, workers subsidize investments in human capital in previous periods by taking
lower wages and receive the benefits of these subsidies in later periods.
In the ESOP only regime, the firm can primarily decide whether to share the assets of the
firm based upon its need for firm specific human capital. As we noted earlier, the share of
profits and the firm is represented by the adoption of ESOPs and determined by the bargaining
power of employees. In the covenants regimes, the firm decides whether to utilize the covenants
not to compete based primarily to make and protect general investments in human capital.
Covenants not to compete may be used with firms which must reveal their intellectual property
with their employee, as a form of investment in general human capital, in order to promote the
development of more intellectual property.
Importantly, the use of covenants not to compete by firms should increase the risk of a
worker. That is, if a worker contracts with a covenants not compete, then they would need to be
compensated for the added risk of a below market wage if they left the present company. Part of
the compensation for the risk may occur through increased firm asset sharing during the first
period contracting phase. Effectively, worker power should increase with the covenants not to
compete regime because the increased risk to workers would require a higher level of asset
sharing and stakeholding in order for them to make these deals.
Therefore, firms that use
covenants not to compete because of need to invest in general human capital or protect
intellectual capital should therefore increase their firm asset sharing devices, such as ESOPs, due
to the increase in worker risk and worker power.
(HI - Rational Hypothesis) The hypothesis, under rational-material perspectives is
that covenants not to compete regime should at least maintain the rate of adoption of ESOPs.
33 1Asset Sharing and Stakeholder Arrangements
Under a distributive conception, the mere creation of economic institutions and property
rights allocate, segment, and distribute powers related to firm asset sharing and employee
stakeholding. In this case, the enforcement of the covenants represents a structural re-allocation
of powers from the previous allocation of powers.
The covenants provide firms with an
alternative to realize and protect their economic interests and realize their economic purposes.
The availability of covenants not to compete increases the alternatives available to protect firm
side interests; alternatives which limit employee powers. It is important to remember that this
reallocation is occurring in a situation of pre-distributed property rights, which involve credit and
liquidity constraints, as well as other distributive features that create inefficient bargaining for
employees and opportunities for moral hazard for firms.
(H2 - Distributive Hypothesis) The distributive hypothesis is that the re-allocationof
powers involved in the enforcement of covenants not to compete reduces the bargaining power
of employees related to firm asset-sharing and employee stakeholding and thus decrease the
adoption of ESOP plans.
Under the distributive conception, we also contend that powers are shaped by how
property rights and economic institutions allocate and segment within property objects and
economic agents. As discussed previously, firms have the ability to unilaterally initiate ESOPs.
ESOPs can be initiated to exclude union workers from the agreement and from participating in
ESOP governance.
(H3 -
Distributive Hypothesis with
Unions) Union bargaining power will not
necessarily have a positive association with ESOP adoption.
The cognitive-cultural
hypothesis proposes
that mechanisms
other
than rational
efficiency mechanisms or quasi-rational distributive mechanisms may be involved in the uptake
34 1Asset Sharing and Stakeholder Arrangements
of economic institutions after a rule change. In this perspective, the legal change encourages
normative or cognitive shift of appropriateness and legitimacy. What is difficult in the present
case is determining the difference between a mere normative shift or a distributive shift.
Therefore, I argue that a normative shift may be best explored in nearby subnational contexts
which are not directly affected by the rules. I employ a regional indicator that captures the effect
of the region surrounding Michigan including, Wisconsin, Ohio, and Illinois.
(H4 - Cognitive-Cultural Hypothesis) The adoption of ESOP in nearby states may be
negatively affected by a shift in covenants not to compete in Michigan.
35 Asset Sharing and Stakeholder Arrangements
0.5 Estimation Strategy
The current research assesses the causal impact of a change in property rights -
the
enforcement of noncompetition provision in Michigan - on the adoption of firm asset sharing
and employee stakeholder arrangements. In the quasi-experimental setting of Michigan's change
in law, I utilize a Difference-in-differences estimator. Difference-in-differences estimators are a
version of fixed effects estimator using more aggregate data (Angrist and Pischke, 2008). The
key identification assumption for Difference-in-differences estimation is parallel trends in the
treated Yj1 and nontreated groups YO such that it is only the treatment which causes the deviation
from the common trend (Angrist and Pischke, 2008). Although treated groups can differ, this
should be captured by the fixed effect for the unit i (Angrist and Pischke, 2008). More formally
E[Y1 (1) - YO(1)ID = 1]
= {E[Y(1)ID = 1] - E[Y(1)ID = 0}
-
{E(Y(O)ID = 1] - E[Y(O)ID = 0]}
where treated groups are D = 1, control groups are D=0, the pre-period t=0, and post-period, t=1.
Card (1990) utilized the Difference-in-differences estimator to model the effect of immigration
on unemployment following the Mariel boatlift from Cuba to Miami.
Additionally, we will
utilize a Difference-in-differences estimator with synthetic control methods (Abadie, Diamond,
and
Hainmueller,
2011).
Synthetic
control
methods
resolve
problems
with
traditional
comparative case studies where control groups are arbitrarily selected. Synthetic control methods
create synthetic control units that are a weighted average of the available control units that
approximate treated units (Abadie, et al 2011).
Additionally, the present research will utilize a count model process with a Difference-indifferences estimator to understand other features of the hypothesized relationships. A key
36 1Asset Sharing and Stakeholder Arrangements
feature of the data set is that the trusts are modeled as count variables and has a panel structure.
Count variables take on non-negative integer values, have no natural upper bound, and have
population members who will be zero (Wooldridge, 2001). Traditional linear OLS models are
generally found not to fit the assumptions of the data generating process for count data4 .
Additionally, OLS models have limitations when used with panel data'.
Therefore, the
methodological response to the count and panel feature of the data is to utilize a poisson process
augmented by a fixed effect strategy. This approach relies on Hausman, Hall, and Griliches
(1984), who have modeled a poisson process in panel data sets, examining the research and
development expenditures of firms and number of patents applied for and received. A poisson
model is used for count data where the response variable is modeled as having a poisson
distribution.
The assumptions for a poisson process include equidispersion,
constant rate, and
independence assumptions. Equidispersion assumes that the conditional variance is equal to the
mean. The constant rate and independence assumptions of the poisson distribution assumes that
event outcomes are independent over time such that when an event occurs it does not affect the
probability that the event occurs in the future. The variance of the poisson distribution is only
equal to the mean if the constant rate assumption and the independent assumptions are true. If
4 In part, because the dependent variable is only zero or greater than zero with count variables, its expected value
conditional on the independent variables X_{i}should be nonnegative for all X_{i}, but the OLS estimator can result
in values of the X_{i} such that the predicted value of dependent variable Y_{i} is negative (Wooldridge, 2002). In
addition, when the dependent variable Y_{i}
can take on the value zero with positive probability, the
transformation of log*Y {i} cannot be used to obtain a model with constant elasticities or constant semi-elasticities
(Wooldridge, 2002).
5 OLS processes assume that the variance of the errors is the same for each unit of analysis by the homoskedasticity
assumption (Castilla, 2007). Moreover, OLS models assume that errors are not correlated across sections or cases
with any of the independent variables (Castilla, 2007). However, cross-sectional and cross-time data are known to
have observations that may not be independent from each other leading to erroneous correlations across panels and
time (Castilla, 2007). While heteroskedasticity does not bias the OLS estimators, it does make the standard errors
inefficient (Castilla, 2007).
37 Asset Sharing and Stakeholder Arrangements
these assumptions are interrupted, then a negative binomial process may be used where the
variance is greater than the mean, traditionally known as overdispersion (Wooldridge, 2002). It is
also possible that data may have a larger frequency of zeros than what is expected from a poisson
or negative binomial processes. In such cases, a zero inflated count process may be helpful
because it assumes that there are multiple processes: one for the count process and one that
produces the additional zeros.
A key opportunity is the panel structure of this data which allows us to examine more
individual level heterogeneity, in this case state level heterogeneity (Cameron and Trivedi,
1989). Here we utilize the poisson fixed effects model instead of the random effects model6 .
Traditionally with count data, the researcher would need to determine which count models poisson, negative binomial, zero-inflated -
best fit the data. However, Cameron and Trivedi
(1989) argue that because using negative binomial and zero inflated models are used to control
for unobserved heterogeneity, these techniques may not be necessary since fixed effect
longitudinal models already control for unobserved heterogeneity (Cameron and Triveldi, 1989).
As such, I will utilize fixed effects with both poisson pooled and fixed effect models.
More
formally, a poisson model where the response variable Yt is modeled as having a poisson
distribution.
Pr(Yjt
=
yjA')=H
6 Generally, the alphai are unknown parameters (Cameron and Trivedi, 1989). It is notable that in count
model, individual specific effects are multiplicative. Fixed effects count data models can be estimated using
maximum likelihood and conditional maximum likelihood, and moment based models (Cameron and Trivedi, 1989).
38 1Asset Sharing and Stakeholder Arrangements
Similar to the previous model, I utilize a regression version of the Difference-indifferences estimator to understand the treatment effect in Michigan. Angrist and Pischke (2002)
highlight the usefulness of a regression version of the Difference-in-differences
particularly as it permits additional controls, covariates, and tests for causality.
estimator,
The Difference-
in-differences estimator thus appears as follows
pg e
"*D +(5*
T+
(D *T) + X&B)
where D is a dummy variable for the state, T is a dummy variable for the time period, and the
coefficient a is the treatment effect that we are measuring.
Notably, Besley and Case (1994) argue that studies that involve state laws should be
attentive to policy endogeneity. Because state policy making is often purposeful and responds to
local conditions within a state, it is necessary to identify and control for these forces in order to
obtain unbiased estimates of a policy (Besley and Case, 1994). However, examining the impact
of non-competes agreements on employee mobility, Marx, Strumsky, and Fleming (2010) argue
that in 1985, Michigan randomly reversed state enforcement of non-compete agreements. They
argue that the bill was an unintentional byproduct of the uniform acts where states align their
provisions with other states. (Marx et al, 2010). The legislative bill contained the Restatement of
Contracts and New York's statute on covenants not to compete. In addition, labor lawyers
highlighted that the repeal of the non-competes agreements was unintentional (Marx et al, 2010).
Moreover, they argue that the Michigan repeal is notable because generally courts merely tighten
or loosen constraints, but the Michigan provision fully reversed the previous policy.
This has two implications for the present research project. First, we argue that the effect
of the Michigan state law is different from more marginal changes in laws, for instance, due to
changes in legal interpretation. We will model the general marginal effect of covenants not to
39 1Asset Sharing and Stakeholder Arrangements
compete, utilizing the Garmaise (2011) index of covenants not to compete, differently than how
we model an exogenous change in Michigan's laws. Notably, the effect of the index will only be
revealed in the pooled poisson model due to the lack of variation across the years in question.
The change in Michigan's laws will be revealed in both the pooled model and fixed effect
version of the count models.
Figure 3 visually represents the negative change in the adoption of ESOP plans in
Michigan during the 1985 to 1987 period. The key question our estimation seeks to resolve is
whether the decrease in the adoption of ESOPs is caused by the change in the enforcement of
covenants not to compete or some other factors. Measuring the effect starting in the year 1985 is
prudent because the establishment of many ESOP agreements occur at the end of the year as
firms try to complete these arrangements before the next year due to the numerous tax benefits.
As such, any change in law that occurs at the beginning of the year will be reflected in plan
adoption by the end of the year.
Tables 4 and 5 display the results from the change in the enforceability of the Michigan
law on covenants not to compete based on a 3.year regime model. Table 5 utilizes an incident
ratio rate formulation. Generally, the quantity of interest in a poisson model is the expected
count. To compare two observations, we must divide the expected count for one scenario to
some other scenario. Generally an incident ratio rate coefficient in Tables 5 and 6 signals that if a
value is above 1 there is a percentage increase in the effect. If the value is below 1, there is a
percentage decrease in the effect.
40 1Asset Sharing and Stakeholder Arrangements
0.6 Results
I utilize the Difference-in-differences models first to test causal effects. Initially the pre
and post data are pooled for the treatment and controls. The broad findings indicate that the
covenants not to compete result in a decrease in the adoptions of ESOP arrangements.
This is
reflected in Table 3a and Figure 5, which show the pre and post difference in Michigan versus
other states. Here it is shown that covenants not to compete generally have a negative effect. I
check these finding by utilizing a synthetic model as demonstrated in Table 3b and Figures 5 and
6. Table 3b contains the collapsed version of the synthetic model which demonstrates only the
pre and post model. It is clear that the synthetic Michigan is increasing in the adoption of ESOP
arrangements; while, the treated unit is decreasing in adoptions.
The non-collapsed, year-by-
year version of the synthetic model in Figure 6 provides a synthetic control and treated unit value
for each year. As shown in Figure 6, the treated unit has a decreasing trend for the 1985 period
when the law was implemented, but not in other periods after the change.
I then model the direct change in Michigan in two ways, remembering that we utilize the
regression version of the Difference-in-differences model estimator to document the state effect,
the time effect, and finally the interaction of the two as the treatment effect of the enforcement
reversal. First, I model the treatment effect as a three year period, covering the period from
1985-1987. This approach is appropriate for multiple reasons. First, the impacts of enforcement
would likely be greatest in the first three years, if we assume that the strength from a change of
law should occur in the first few years after a change. Second, there was a significant change
related to the use of ESOPs as an option to resist hostile takeover culminating in the Delaware
Polaroid case in 1989.
Third, compiling the effects over the three year period may start to
resolve some of problems indicated by Bertrand et al (2002). Table 4 documents the coefficients
411 Asset Sharing and Stakeholder Arrangements
from the direct poisson models. Table 5 and 6 utilize the incident ratio rates (IRR) as described
previously.
In addition, I model the change as yearly effects. This method pennits us to
understand the treatment effects of the law over each year. Table 6 displays the result from the
change in enforceability based on a year by year model, similarly utilizing an incident ratio rate
formulation.
The majority of the models indicate that Michigan indicator and the indicator for the time
period 1985-1987 generally have a positive effect on the adoption of broad-based stock
ownership.
The effect of the regime time period is displayed in the row marked "Time 1985-
1987" in Tables 4 and 5.
These effects are also seen in the yearly model for Michigan and 3
year model for Michigan.
Yet, on the other hand, the Difference-in-differences
estimators labeled "Michigan
Treatment" signal that, while holding all other variables constant, the enforceability of the
covenants not to compete had a significant negative effect upon the adoption of employee stock
ownership plans. This negative effect reads as negative in Table 4 and a value less than one in
Table 5. Generally, the pooled poisson model has a slightly lower significance than the fixed
effect models. They permit the null hypothesis only to be rejected at the .10 level. Whereas, the
poisson fixed effect and negative binomial fixed effect models allows the null hypothesis to be
rejected below the .01 level.
To facilitate interpretation, we utilize the incident ratio rate estimates in Table 6. In Table
5 Model 6, we interpret the impact of the Michigan enforcement/treatment as reducing the
number of adoptions by 37%, holding all other variables constant.
In Table 6 Model 2, we
interpret the yearly effect as reducing the number of adoptions by 45% in 1985. The effect of the
law increases the effect size from 1985 to 1986. Using the incident ratio rate, we find a 22%
42 1Asset Sharing and Stakeholder Arrangements
reduction for 1986 and a 63% reduction for 1987 with a discrete change in enforcement. By
1988, the effect size had changed from negative to positive, indicating an increase of 8%.
This
aligns with our sensibility that other factors were relevant during the later period.
Next, we examine the Garmaise (2011) index on non-competes agreements by state, as
displayed with the gray bar reading "Covenants".
Effectively, this is the marginal effect of legal
variation in non-competes agreement enforcement.
As signaled across all pooled models in
Tables 4 and 5, the marginal effect of increasing enforcement regime is an increase in the
adoption of broad-based stock ownership. This is indicated by a positive value in Table 4 and a
value over 1 in Table 5. This effect is only significant in the pooled models with controls, but
signals a sixteen percent increase in the adoption of ESOPs utilizing the non-competition index.
This effect contravenes theoretical expectations.
Columns 5 and 6 in Tables 5 and 6 include controls for regional effects. This includes
Wisconsin, Illinois, and Ohio. We model this regional effect utilizing a similar Difference-indifferences estimator. Utilizing the pooled estimator in Colunm 5, we find a generally positive
effect of the region on ESOP adoption. However, we find a similar negative effect of the region
during the treatment period. The treatment effect signals a reduction of about 19% for the region,
which is smaller than the effect for solely Michigan.
Generally, these estimates indicate that firms are less likely to adopt these firm assetsharing and employee stakeholder arrangements when the noncompetition law in Michigan was
enforced. This aligns with both traditional bargaining power theories and distributive theories of
property rights and economic institutions. On the other hand, the index for non-competes
agreements contravenes theoretical expectations for bargaining power.
43 1Asset Sharing and Stakeholder Arrangements
These findings also suggest that a change of law in Michigan affected the region more
generally. This regional effect has two possible implications. First, despite our controls, there
may be a common unobserved variable that affected the region during this period. Alternately,
the change in enforcement affected not merely the state in question, but also the region.
Previously, we proposed that this mechanism for the region may have occurred through the shifts
in cultural-cognitive institutions, which include normative expectations, mimetic pressures, and
cognitive processes. This could have occurred through multiple channels. For instance, media
publicity of the change in enforcement may have signaled that other states may follow
Michigan's direction, thus employers reduced their use of ESOP mechanisms. We believe the
second interpretation is more appropriate given that the size of the effect for the region is much
smaller.
The findings for unions confirm the predictions derived from distributive conceptions of
property rights and economic institutions and the segmented allocations of power. In our
bivariate analysis demonstrated in Table 6, unions have a positive relationship with the adoption
of broad-based stock ownership that is significant at the below the .10 level of the p-value.
However, this relationship reverses once we control for other factors. Tables 7 and 8 present the
results for the effect of unionization on the adoption of employee stakeholder arrangements.
Only the pooled models demonstrate a significant effect for unionization on firm-asset sharing.
None of the fixed effect models demonstrate a significant effect for unionization.
Generally,
across all models, the relationship between unionization and employee stakeholder arrangements
was negative. This parallels other findings for this time period. The quantity of interest decreases
by approximately 1% across all models with a one factor decrease in unionization.
44 1Asset Sharing and Stakeholder Arrangements
This
finding suggests that unionization
confirms the hypotheses emerging from
distributive conception of property rights and economic institutions and how they distribute
power.
In addition to our primary hypotheses, I also sought to evaluate a number of hypotheses
that emerged from the literature on human capital and profit sharing firms. Aligning with
theoretical expectations, increases in the population of persons with more than six years of
college had a positive effect on the adoption of broad-based stock ownership.
However, a
significant positive relationship only appeared in the pooled model. The fixed effect models
contained both negative and positive results, but none of these were significant. Interestingly, an
increase in the percentage of persons with four year degrees showed a negative effect across all
models. This effect was only significant in the pooled poisson model with controls at the ten
percent level of the p-value. The variable percent professional also maintained a negative
relationship across both of the fixed effect models, but this finding was similarly not significant.
As such, the estimates are not clear as to what role human capital generally plays on the adoption
of employee stakeholder arrangements, but does suggest that more than six years of college has a
positive effect on adoption.
An additional area of inquiry is related to the population of businesses. The coefficient
for businesses was negative, but extremely small. When this coefficient is converted to the
incident rate ratio it becomes unity. This suggests that increases in the population of businesses
had a marginally negative impact on firm asset-sharing. This was consistent across both poisson
fixed effect models. This supports the theoretical expectation that increasing oligopoly would
decrease the adoption of broad-based ownership arrangements.
45 1Asset Sharing and Stakeholder Arrangements
0.7 Conclusions
The present research has explored how a change in economic institutions and property
rights affects the adoption of firm asset sharing and employee stakeholder arrangements. It has
sought to look beyond theories of human capital and bargaining power to understand the more
subtle ways property rights and economic institutions shape economic processes and decisionmaking.
Grounded in theories of how institutions constitute markets, this analysis utilized
arguments of economic institutions that encompassed rationalist, distributive, and cognitivecultural conceptions.
Part of the puzzle in the literature has been the differential effect of unions on the
adoption of particular employee stakeholder arrangements, such as ESOPs. Under a classical
bargaining power analysis, unions should increase the adoption of firm asset sharing and
employee stakeholder arrangements which has contravened empirical evidence. We employ a
distributive
conception which
proposes that property
differentially allocate, segment, and distribute powers.
rights and
economic
institutions
These features arise from the pre-
distributed, pre-configured, and power-allocative features of property rights and economic
institutions. Distributive conceptions suggest that economic institutions allocate and distribute
powers through the creation of property objects and economic agents. Therefore, unlike classical
conceptions, power is not simply a sum function, but actually may be segmented and distributed
differently by property rights and economic institutions.
These differences produce varied
configurations of interests and varied relationships with other property objects.
Specifically, we examine the impacts of covenants not to compete and unions. Consistent
with the traditional bargaining power view and a distributive conception, I find that a discrete
state law change of enforcement of noncompetition agreements matches theoretical expectations.
46 1Asset Sharing and Stakeholder Arrangements
These agreements curtailed the previous allocation of powers provided by ESOPs, by creating
alternative property objects.
. The specific estimates suggest that Michigan's law resulted in
almost a third less ESOP adoptions than prior to the law.
Second, consistent with our hypotheses regarding the distributed and segmented impact
of property rights and economic institutions on allocations of power, we find in panel data that
union density does not predict the adoption of ESOPs as employee stakeholder arrangements.
This finding aligns well with previous research on the relationship of unions with employee
stakeholder arrangements such as ESOPs. Notably, the covenants may have not only affected the
state in question, but the region more generally. We argue that this supports cognitive-cultural
theories of economic institutions affecting the adoption of firm asset sharing arrangements. We
presume that Michigan state law does not have jurisdiction over firms in the region.
However, I discover one further question related to bargaining power.
The present
findings indicate that the marginal enforcement of covenants not to compete, reflected in the
Garmais (2011) index of the strength of noncompetition enforcement by state, may have a
positive relationship with the adoption of employee ownership. This finding suggests that policy
endogeneity may be a real issue when examining marginal variations in state law that may be
caused by gradations in legal interpretation, rather than discrete changes in enforcement.
A key limitation of the present research is its level of analysis. I examined these issues at
the level of states. This permits us to measure larger global effects, for example educational
levels, unemployment, capital availability, or other effects in the spirit of comparative case
studies. However, more localized or micro-firm data may be more helpful in truly understanding
these effects as they are likely not the same across areas in a state, for example between urban,
suburban, and rural areas. I have not controlled for variations in the national or regional business
47 1Asset Sharing and Stakeholder Arrangements
cycle which might have a significant effect on the outcomes if they are drivers. Finally, in future
iterations, it may be useful to formalize the informal conversational game model used here.
Nonetheless, this research seems to support broadly the distributive conception that
changes in economic institutions and property rights affects the distribution of powers and thus
the adoption of firm asset sharing and employment stakeholder arrangements, such as ESOPs.
Moreover, it supports the conception that these allocations of power by economic institutions and
property rights are distributed and segmented such that not all bargaining power has the same
effect on the adoption of firm asset-sharing and employer stakeholder arrangement. Future
research may seek to be more attentive to pathways by which specific features of economic
institutions and property rights distribute powers to employees and other economic actors as well
as their affects in varying contexts
If employee stakeholding
and firm-asset
sharing arrangements
are considered
a
substantive indicator of structural variation in the firm, the present research may have important
implications for considering the relationship between the labor, property rights, and corporate
structure, and inequality outcomes. Additionally, these findings may inform theory about what
legal factors influence firms to choose high road or low road options in wage and benefit
structuring. Finally, these findings may suggest the direction that policy-makers may not want to
pursue in order to encourage the adoption of labor stakeholder arrangements and firm asset
sharing with employees.
48 1Asset Sharing and Stakeholder Arrangements
0.8 Bibliography
Abadie, A., D. Drukker, J. L. Herr, and G. W. Imbens. 2004. "Implementing matching estimators
for average treatment effects in Stata." Stata journal 4:290-311..
Abadie, Alberto, Alexis Diamond, and Jens Hainmueller. 2011. "Synth: An R Package for
Synthetic Control Methods in Comparative Case Studies." Journal of Statistical Software
42(13)..
Acemoglu, D. 1999. "Lecture Notes for Graduate Labor Economics, 14.661." Massachusetts
Institute of Technology, Cambridge..
Acemoglu, Daron, and Jorn-Steffen Pischke. 1998. The Structure of Wages and Investment in
General Training. National Bureau of Economic Research..
Acemoglu, Daron, and Jorn-Steffen Pischke. 1999. "Beyond Becker: Training in Imperfect
Labour Markets." The Economic Journal 109(453):112-42.
Acemoglu, Daron, Simon Johnson, and James A. Robinson. 2005. "Chapter 6 Institutions as a
Fundamental Cause of Long-Run Growth." Pp. 385-472 in Handbook of Economic
Growth, vol. Volume 1, Part A, edited by Philippe Aghion and Steven N. Durlauf.
Elsevier.
Acemoglu, Daron. 1997. "Training and Innovation in an Imperfect Labour Market." The Review
of Economic Studies 64(3):445-64.
Aguilera, Ruth V., and Gregory Jackson. 2003. "The Cross-National Diversity of Corporate
Governance: Dimensions and Determinants." The Academy of Management Review
28(3):447-465.
Aldatmaz, Serdar, Paige Parker Ouimet, and Edward Dickersin Van Wesep. 2011. "The Option
to Quit: The Effect of Employee Stock Options on Turnover." SSRN eLibrary.
Angrist, Joshua D., and Jbrn-Steffen Pischke. 2008. Mostly Harmless Econometrics: An
Empiricist's Companion. Princeton University Press..
Anon. n.d. "Review: Jacob Hacker and Paul Pierson - Winner-Take-All Politics -
Crooked
Timber.".
Aoki, M. 1980. "A model of the firm as a stockholder-employee cooperative game." The
American Economic Review 600-610.
49 1Asset Sharing and Stakeholder Arrangements
Aoki, M. 1982. "Equilibrium growth of the hierarchical firm: Shareholder-employee cooperative
game approach." The American Economic Review 1097-1110.
Aoki, M. 1984. The co-operative game theory of the firm. Clarendon Press Oxford.
Autor, David. 2001. "Wiring the Labor Market." Journal of Economic Perspectives 15(1):25-40.
Bagchi, A. 2007. "Varieties of Employee Ownership: Some Unintended Consequences of
Corporate Law and Labor Law." U. Pa. J. Bus. & Emp. L. 10:305.
Becker, G. S. 1975. "Front matter, Human Capital: A Theoretical and Empirical Analysis, with
Special Reference to Education." Pp. 22-0 in Human Capital: A Theoretical and
Empirical Analysis, with Special Reference to Education, 2nd ed. NBER.
Becker, Gary S. 1962. "Investment in Human Capital: A Theoretical Analysis." Journal of
Political Economy 70(5):9-49.
Bertrand, M., E. Duflo, and S. Mullainathan. 2002. How much should we trust differences-indifferences estimates? National Bureau of Economic Research.
Besley, T., and A. Case. 1994. Unnatural experiments? Estimating the incidence of endogenous
policies. National Bureau of Economic Research.
Blair, M. M. 1995. Ownership and control: Rethinking corporate governance for the twenty-first
century. Brookings Institution Press
Blair, M. M., and T. A. Kochan. 2000. The new relationship: Human capital in the American
corporation. Brookings Institution Press.
Blair, M.M. 1999. "Firm-specific human capital and theories of the firm." Employees and
corporate governance 58:290.
Blair, Margaret M., and Mark J. Roe. 1999. Employees and corporate governance. Brookings
Institution Press.
Blasi, J.R., D.L. Kruse, and R.B. Freeman. 2010. Epilogue (and Prologue). University of
Chicago Press.
Breyer, Stephen G., and Stephen G. Breyer. 2009. Regulation and Its Reform. Harvard
University Press.
Buchele, R., D.L. Kruse, L. Rodgers, and A. Scharf. 2010. Show Me the Money: Does Shared
Capitalism Share the Wealth? University of Chicago Press.
Budros, A. 1997. "The new capitalism and organizational rationality: The adoption of
downsizing programs, 1979-1994." Social Forces 76(1):229-250.
50 1Asset Sharing and Stakeholder Arrangements
Cahuc, P., and A. Zylberberg. 2004. Labor economics. The MIT Press
Cameron, A. C., and P. K. Trivedi. 1986. "Econometric models based on count data.
Comparisons and applications of some estimators and tests." Journal of applied
econometrics 1(1):29-53..
Cameron, A. C., and P. K. Trivedi. 1998. Regression analysis of count data. Cambridge
University Press.
Campbell, John L., and Leon N. Lindberg. 1990. "Property Rights and the Organization of
Economic Activity by the State." American Sociological Review 55(5):634.
Carberry, E.J. 2010. Who Benefits from Shared Capitalism? The Social Stratification of Wealth
and Power in Companies with Employee Ownership. University of Chicago Press.
Card, David. 1990. "The Impact of the Mariel Boatlift on the Miami Labor Market, 43 INDUS.
& LAB." Rel. Rev 245:250.
Carruthers, Bruce G., and Laura Ariovich. 2004. "The Sociology of Property Rights." Annual
Review of Sociology 30(1):23-46.
Castilla, E. J. 2007. Dynamic analysis in the social sciences. Academic.
Clegg, Stewart. 1981. "Organization and Control." Administrative Science Quarterly 26(4):545.
Coff, Russell W. 1999. "When Competitive Advantage Doesn't Lead to Performance: The
Resource-Based View and Stakeholder Bargaining Power." Organization Science
10(2):119-33.
De Soto, H. (2000). The mystery of capital: Why capitalism succeeds in the West and fails
everywhere else.
DiMaggio, P. J, and W. W Powell. 1983. "The iron cage revisited: Institutional isomorphism and
collective rationality in organizational fields." American sociological review 48(2):147-
160.
Dobbin, F., J.R. Sutton, J.W. Meyer, and R. Scott. 1993. "Equal opportunity law and the
construction of internal labor markets." American Journal of Sociology 396-427.
Dobbin, Frank, and Timothy J. Dowd. 2000. "The Market That Antitrust Built: Public Policy,
Private Coercion, and Railroad Acquisitions, 1825 to 1922." American Sociological
Review 65(5):631-657.
Dow, G. K. 2003. Governing the firm: workers' control in theory and practice. Cambridge Univ
Pr.
51 1Asset Sharing and Stakeholder Arrangements
Edelman, L.B., and M.C. Suchman. 1997. "The legal environments of organizations." Annual
review of sociology 479-515.
Edelman, Lauren B. 1990. "Legal Environments and Organizational Governance: The Expansion
of Due Process in the American Workplace." American Journal of Sociology
95(6):1401-1440..
Eisenhardt, K. M, and J. A Martin. 2003. Dynamic capabilities: what are they? Blackwell,
Malden, MA.
Esping-Andersen, G. 1999. Social foundations of postindustrial economies. Oxford University
Press Oxford.
Esping-Andersen, Gosta. 1990. "The Three Worlds of Welfare Capitalism.".
Estevez-Abe, M., Iversen, T., & Soskice, D. (2001). Social protection and the formation of skills:
a reinterpretation of the welfare state. Varieties of capitalism: The institutional
foundations of comparative advantage, 145-83.
Estrin, Saul, Paul Grout, Sushil Wadhwani, S. J. Nickell, and Mervyn King. 1987. "ProfitSharing and Employee Share Ownership." Economic Policy 2(4):13-62.
Evans, David S., and Boyan Jovanovic. 1989. "An Estimated Model of Entrepreneurial Choice
under Liquidity Constraints." The Journal of Political Economy 808-27.
Evans, David S., and Linda S. Leighton. 1989. "Why Do Smaller Firms Pay Less?" Journal of
Human Resources 299-318.
Farber, H. S. 2005. "Union membership in the United States: The divergence between the public
and private sectors." WORKING PAPER-PRINCETON UNIVERSITY INDUSTRIAL
RELATIONS SECTION 503.
Fligstein, N. 2001. "Social skill and the theory of fields." Sociological theory 19(2):105-125.
Fligstein, N., and J. Choo. 2005. "Law and corporate governance." Annu. Rev. Law Soc. Sci.
1:61-84.
Fligstein, Neil, and Jennifer Choo. 2005. "Law and Corporate Governance." Annual Review of
Law and Social Science l(1):61-84..
Fligstein, Neil, and Roberto M. Fernandez. 1988. "Worker Power, Firm Power, and the Structure
of Labor Markets." Sociological Quarterly 29(1):5-28.
Fligstein, Neil. 1996. "Markets as Politics: A Political-Cultural Approach to Market
Institutions." American Sociological Review 61(4):656.
52 1Asset Sharing and Stakeholder Arrangements
Freeman, R. B., and J. L. Medoff. 1984. "What Do Unions Do." Indus. & Lab. Rel. Rev. 38:244.
Freeman, R., and A. Dube. 2000. "Shared compensation systems and decision making in the US
job market." Draft, Harvard University Department of Economics.
Freeman, R.B., J.R. Blasi, and D.L. Kruse. 2010. Introduction to" Shared Capitalism at Work:
Employee Ownership, Profit and Gain Sharing, and Broad-based Stock
Options."University of Chicago Press.
Freeman, S. F., and M. Knoll. 2008. "S Corp ESOP Legislation Benefits and Costs: Public
Policy and Tax Analysis." Organizational Dynamics Working Papers 4.
Furubotn, Eirik Grundtvig, and Rudolf Richter. 2005. Institutions and Economic Theory: The
Contribution of the New Institutional Economics. University of Michigan Press.
Garmaise, Mark J. 2011. "Ties that Truly Bind: Noncompetition Agreements, Executive
Compensation, and Firm Investment." Journal of Law, Economics, and Organization
27(2):376-425.
General Accounting Office. 1986. GAO Employee Stock Ownership Plans: Benefits and Costs
of ESOP Tax Incentives for Broadening Stock ownership.
Gibbons, Robert. 2005. "Four Formal(izable) Theories of the Firm?" Journal of Economic
Behavior & Organization 58(2):200-245.
Gittleman, M., and E. N Wolff. 2004. "Racial differences in patterns of wealth accumulation."
Journal of Human Resources 39(1):193.
Gospel, H., and A. Pendleton. 2003. "Finance, corporate governance and the management of
labour: a conceptual and comparative analysis." British Journal of Industrial Relations
41(3):557-582.
Grossman, Sanford J., and Oliver D. Hart. 1986. "The Costs and Benefits of Ownership: A
Theory of Vertical and Lateral Integration." Journal of Political Economy 94(4):691.
Grout, Paul A. 1984. "Investment and Wages in the Absence of Binding Contracts: A Nash
Bargaining Approach." Econometrica 52(2):449-460..
Hacker, Jacob S., and Paul Pierson. 2010. "Winner-Take-All Politics: Public Policy, Political
Organization, and the Precipitous Rise of Top Incomes in the United States." Politics &
Society 38(2):152-204.
Hacker, Jacob S., and Paul Pierson. 2011. Winner-Take-All Politics: How Washington Made the
Rich Richer--and Turned Its Back on the Middle Class. Simon and Schuster.
53 1Asset Sharing and Stakeholder Arrangements
Hansmann, H. 1989. "When Does Worker Ownership Work-ESOPs, Law Firms,
Codetermination, and Economic Democracy." Yale LJ 99:1749.
Hansmani, H. 1999. "Cooperative firms in theory and practice." LTA 48(4):387-403.
Hansmann, H., and R. Kraakman. 2000. "End of History for Corporate Law, The." Geo. LJ
89:439.
Hart, Oliver D., and John Moore. 1990. Property Rights and the Nature of the Firm. Cambridge,
Mass. : Massachusetts Institute of Technology.
Hart, Oliver. 1989. "An Economist's Perspective on the Theory of the Firm." Columbia Law
Review 89(7):1757.
Hausman, J. A., B. H. Hall, and Z. Griliches. 1984. Econometric models for count data with an
application to the patents-R&D relationship. National Bureau of Economic Research
Cambridge, Mass., USA.
Holtz-Eakin, Douglas, David Joulfaian, and Harvey S. Rosen. 1993. Sticking It out:
Entrepreneurial Survival and Liquidity Constraints. National Bureau of Economic
Research..
Hurst, Erik, and Annamaria Lusardi. 2004. "Liquidity Constraints, Household Wealth, and
Entrepreneurship." Journal of political Economy 112(2):319-47.
Inderst, R., and H. M0ller. 2007. "Safeguarding specific human capital investments." Unpub)
lished working paper, New York University.
Jou, J., and T. Lee. 2004. "Bargaining, Profit-Sharing and Irreversible Investment Decisions In
An Oligopoly."
Kaarsemaker, E., A. Pendleton, and E. Poutsma. 2009. "Employee Share Ownership Plans: A
Review."
Kalleberg, A. L. 2011. "Good jobs, bad jobs." The rise of polarized and precarious employment
systems in the United States, 1970s to 2000s. Russell Sage Foundation, New York..
Kedia, S., and S. Rajgopal. 2009. "Neighborhood matters: The impact of location on broad based
stock option plans." Journal of Financial Economics 92(1):109-127.
Kennedy, David. 2011. "Accounting, Economics, and Law."
Kenworthy, Lane, and Jonas Pontusson. 2005. "Rising Inequality and the Politics of
Redistribution in Affluent Countries." Perspectives on Politics 3(03):449-7 1.
54 1Asset Sharing and Stakeholder Arrangements
Kerckhoff, Alan C. 1995. "Institutional Arrangements and Stratification Processes in Industrial
Societies." Annual Review of Sociology 21:323-347.
King, Miriam et al. n.d. Minneapolis: University of Minnesota, 2010.
Klein, Benjamin, Robert G. Crawford, and Armen A. Alchian. 1978. "Vertical Integration,
Appropriable Rents, and the Competitive Contracting Process." Journal of Law &
Economics 21:297.
Kochan, T. A. 2011. "RESOLVING AMERICA'S HUMAN CAPITAL PARADOX: A
PROPOSAL FOR A JOBS COMPACT."
Kochan, T. A., R. B. McKersie, and J. Chalykoff. 1986. "The effects of corporate strategy and
workplace innovations on union representation." Industrial and Labor Relations Review
487-501.
Kochan, Thomas A., and Saul A. Rubinstein. 2000. "Toward a Stakeholder Theory of the Firm:
The Saturn Partnership." Organization Science 11(4):367-86.
Kochan, Thomas A., Harry Charles Katz, and Robert Bruce McKersie. 1986. The
Transformation of American Industrial Relations. Cornell University Press. Retrieved.
Kochan, Thomas A., Harry Charles Katz, and Robert Bruce McKersie. The transformation of
American industrial relations. Cornell University Press, 1986.
Kruse, Douglas L. 1996. "Why Do Firms Adopt Profit-Sharing and Employee Ownership
Plans?" British Journal of Industrial Relations 34(4):515-538.
Kuklys, W., and I. Robeyns. 2005. "Sen's capability approach to welfare economics." Amartya
Sen's Capability Approach 9-30.
Lupu, Noam, and Jonas Pontusson. 2011. "The Structure of Inequality and the Politics of
Redistribution." American Political Science Review 105(02):316-36.
Marx, Karl. n.d. The Economic and Philosophic Manuscripts of 1844 and the Communist
Manifesto. Prometheus Books.
Marx, M. 2011. "The Firm Strikes Back Non-compete Agreements and the Mobility of
Technical Professionals." American Sociological Review 76(5):695-712.
Marx, M., D. Strumsky, and L. Fleming. 2010. "Mobility, skills, and the Michigan non-compete
experiment." Management Science 55(6):875.
Marx, M., J. Singh, and L. Fleming. 2010. "Regional Disadvantage? Non-compete agreements
and brain drain."
55 Asset Sharing and Stakeholder Arrangements
McCarth, J., P. Voos, A.E.Eaton, D.L Kruse, and J.R. Blasi. 2011. "Solidarity and Sharing:
Unions and Shared Capitalism" in Who Benefits from Shared Capitalism? The Social
Stratification of Wealth and Power in Companies with Employee Ownership. Edited by
E.J. Carberry. University of Chicago Press.
McElrath, Roger G., and Richard L. Rowan. 1992. "The American Labor Movement and
Employee Ownership: Objections to and Uses of Employee Stock Ownership Plans."
Journal of Labor Research 13(1):99-119.
Mchugh, Patrick P., Joel Cutcher-Gershenfeld, and Michael Polzin. 1999. "Employee stock
ownership plans: union influence and stakeholder interests." Economic and Industrial
Democracy 20(4):535-560.
Meyer, J. W, and B. Rowan. 1977. "Institutionalized organizations: Formal structure as myth and
ceremony." The American journal of sociology 83(2):340-363.
Moretto, M, and G Rossini. 1995. "The Shut-Down Option and Profit Sharing." Journal of
Comparative Economics 21(2):154-86.
Moretto, Michele, and Gianpaolo Rossini. 1996. "Profit Sharing Regulation and Repeated
Bargaining with a Shut-down Option." Economic design 2(1):339-68.
Morgan, D., B. Rime, and P. Strahan. 2003. Bank integration and state business cycles. National
Bureau of Economic Research.
Mouw, T., and A. L. Kalleberg. 2010. "Occupations and the Structure of Wage Inequality in the
United States, 1980s to 2000s." American Sociological Review 75(3):402-43 1..
NCEO. 2011. "A Comprehensive Overview of Employee Ownership."
Pendleton, Andrew, and Andrew Robinson. 2011. "Employee share ownership and human
capital development: Complementarity in theory and practice." Economic and Industrial
Democracy 32(3):439 -457.
Perrow, Charles. 1988. Markets, Hierarchies and Hegemony. Harvard University Press.
Pynnonen, B. D. 1994. "Ohio and Michigan Law on Postemplyoment Covenants Not to
Compete." Ohio St. LJ 55:215.
Rajan, Raghuram G., and Luigi Zingales. 1998. "Power in a Theory of the Firm." The Quarterly
Journal of Economics 113(2):387-432.
Roberts, D., and E. Van den Steen. 2000. "Shareholder interests, human capital investment and
corporate governance."
56 1Asset Sharing and Stakeholder Arrangements
Robinson, Andrew M, and Hao Zhang. 2005 "Employee Share Ownership: Safeguarding
Investments in Human Capital." British Journal of Industrial Relations 43(3):469-488..
Rousseau, D.M., and Z. Shperling. 2003. "Pieces of the action: Ownership and the changing
employment relationship." Academy of Management Review 28(4):553-570.
Salais, R. 2003. "Work and welfare: toward a capability approach." Governing work and welfare
in a new economy: European and American experiments 315-44.
Schneper, W. D., and M. F. Guillen. 2004. "Stakeholder rights and corporate governance: A
cross-national study of hostile takeovers." Administrative Science Quarterly 49(2):263-
295. Retrieved June 22, 2012.
Selznick, Philip. 1996. "Institutionalism 'Old' and 'New'." Administrative Science Quarterly
41(2):270-277.
Sen, A. 1999. Development as freedom. Oxford University Press.
Shapiro, Thomas M., and Melvin L. Oliver. 2006. Black Wealth/white Wealth: A New
Perspective on Racial Inequality. New York, NY: Routledge.
Sorensen, A. 2005. "Foundations of a rent-based class analysis." Approaches to class analysis
119-151.
Soskice, David, Margarita Estevez-Abe, and Torben Iversen. 2001. "Social Protection and the
Formation of Skills: A Reinterpretation of the Welfare State." Pp. 145-83 in Varieties of
capitalism: the institutional foundations of comparative advantage, edited by David
Soskice and Peter Hall. Oxford: Oxford University Press.
Spilerman, Seymour. 2000. "Wealth and Stratification Processes." Annual Review of Sociology
26:497-524.
Sutton, John R., and Frank Dobbin. 1996. "The Two Faces of Governance: Responses to Legal
Uncertainty in U.S. Firms, 1955 to 1985." American Sociological Review 61(5):794811.
Sutton, John R., Frank Dobbin, John W. Meyer, and W. Richard Scott. 1994. "The Legalization
of the Workplace." American Journal of Sociology 99(4):944-971.
Swedberg, Richard. 2003. "The Case for an Economic Sociology of Law." Theory and Society
32(1):1-37.
57 1Asset Sharing and Stakeholder Arrangements
Swenson, Peter. 2011. "Bringing Capital Back In, or Social Democracy Reconsidered: Employer
Power, Cross-Class Alliances, and Centralization of Industrial Relations in Denmark and
Sweden." World Politics 43(04):513-44.
Swenson, Peter. 2011. "Bringing Capital Back In, or Social Democracy Reconsidered: Employer
Power, Cross-Class Alliances, and Centralization of Industrial Relations in Denmark and
Sweden." World Politics 43(04):513-44.
Teece, D. J. 2009. Dynamic capabilities and strategic management: organizing for innovation
and growth. Oxford University Press, USA.
Teece, David J., Gary Pisano, and Amy Shuen. 1997. "Dynamic Capabilities and Strategic
Management." Strategic Management Journal 18(7):509-533.
Thelen, K. 1999. "Historical institutionalism in comparative politics." Annual review of political
science 2(1):369-404.
Tomaskovic-Devey, D., and K.H. Lin. 2011. "Income Dynamics, Economic Rents, and the
Financialization of the US Economy." American Sociological Review 76(4):538-559.
Wallerstein, M. 1999. "Wage-setting institutions and pay inequality in advanced industrial
societies." American Journal of Political Science 649-680. Retrieved June 22, 2012.
Wallerstein, M., and B. Western. 2000. "Unions in decline? What has changed and why."
Annual Review of Political Science 3(1):355-377.
Weber, Max. 1978. Economy and Society: An Outline of Interpretive Sociology. University of
California Press.
Weber, Max. 2008. The Protestant Ethic and the Spirit of Capitalism with Other Writings on the
Rise of the West. 4th ed. edited by Stephen Kalberg. Oxford University Press, USA.
Weeden, K. A. 2002. "Why Do Some Occupations Pay More than Others? Social Closure and
Earnings Inequality in the United States 1." American Journal of Sociology 108(1):55-
101.
Weeden, K. A., Y. M. Kim, M. Di Carlo, and D. B. Grusky. 2007. "Social class and earnings
inequality." American Behavioral Scientist 50(5):702-736. Retrieved June 22, 2012.
Weeden, K.A. 2002. "Why do some occupations pay more than others? Social closure and
earnings inequality in the United States." ajs 108(1):55-101.
Western, B., and J. Rosenfeld. 201 la. "Unions, Norms, and the Rise in US Wage Inequality."
American Sociological Review 76(4):513-537.
58 | Asset Sharing and Stakeholder Arrangements
Western, Bruce, and Kieran Healy. 1999. "Explaining the OECD Wage Slowdown Recession or
Labour Decline?" European Sociological Review 15(3):233-49.
Williamson, Oliver E. 1971. "The Vertical Integration of Production: Market Failure
Considerations." American Economic Review 61(2):112-23.
Wooldridge, J. M. 2001. Econometric analysis of cross section and panel data. MIT press.
Wright, Erik Olin. 2000. "Class, Exploitation, and Economic Rents: Reflections on Sorenson's
'Sounder Basis' Erik Olin Wright." American Journal of Sociology 105(6):1559-71.
Zingales, L. 1998. Corporate Governance, in the New Palgrave Dictionary of Economics and the
Law, P. Newman Ed. Macmillan, New York, NY.
59 1Asset Sharing and Stakeholder Arrangements
0.9 APPENDIX:
Figure 1: ESOP Trust Adoption by Number of Adoptions per State per Year
0D
LO
V-
CD
0)
0)
LO
II
0
5
10
ESOP Trusts by State and Year
60 1Asset Sharing and Stakeholder Arrangements
15
Figure 2: ESOP Trust Adoption by Year All States
ESOP Trust Adoption by Year
LO
1981
1982
1983
1984
1985
excludes outside values
61| Asset Sharing and Stakeholder Arrangements
1986
1987
1988
1989
1990
Table 1: Plan Adoptions Per Year by State
Plan Adoptions
Freq.
Percent
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
138
111
66
49
40
17
17
18
13
11
5
7
2
4
1
1
27.6
22.2
13.2
9.8
8
3.4
3.4
3.6
2.6
2.2
1
1.4
0.4
0.8
0.2
0.2
Total
Mean
Min
Max
Standard
Deviation
500
2.596
0
15
62 1Asset Sharing and Stakeholder Arrangements
2.989759
Figure 3: ESOP/ Trust Adoption by Year in Michigan
ESOP/Trust Adoption by Year in Michigan
CO
N%
1980
1982
63 1Asset Sharing and Stakeholder Arrangements
1984
year
1986
1988
Table 2: Bivariate Analysis
ESOP
Adoption
1
ESOP Adoption
index*
Index*
0.0642
1
0.1517
* Garmaise Index of Enforcement of Non Competition Agreements by
State
64 1Asset Sharing and Stakeholder Arrangements
Table 3a: Differences in Differences Estimates of the
Effect of Covenants not to Compete
Pre-Period
(1)
Post-Period
(2)
Difference
(3)
Treated
21
16
-5
Control
5.755
8.95
3.195
5
7.05
-8.195
Difference
Table 3b: Differences in Differences Estimates of the Effect of Covenants not to
Compete on Michigan and Synthetic Michigan
Pre-Period
Post-Period
Difference
(3)
(2)
(1)
Treated
21
16
-5
Control
12.906
19.167
6.21
5
-3.167
-11.261
Difference
65 1Asset Sharing and Stakeholder Arrangements
Figure 5: Basic Differences and Differences (Accompanies Table 3a)
Michigan with 3 years Pre and 3 Years Post
o
0
.2
.6
.4
Pre and Post Years
Michigan
66 1Asset Sharing and Stakeholder Arrangements
.8
Other States
Figure 6a: Synthetic Control (Accompanies Table 3b)
Michigan with 3 years Pre and 3 Years Post
C\J
0
.2
.6
.4
Pre and Post Years
Michigan
67 1Asset Sharing and Stakeholder Arrangements
.8
Synthetic Michigan
Figure 6b: Synthetic Control Years Not Collapsed (Accompanies Table 3b)
00
CV
198
185196
183194
yea
trae/nt
-----
68 1Asset Sharing and Stakeholder Arrangements
snhtccnrlui
98
Table 4: Poisson Pooled and Fixed Effect Regressions
(2)
(1)
Models
(3)
(4)
Fixed Effect
Ef
ixe
F
Conrith
t
(5)
(6)
Pooled Poisson
Pool
w iontoal
Fixed Effect
Poisson with
Controls and
Regional Controls
Pooled Poisson
Fixed Effect
Poisson
Pooled Poisson
With Control
0.256**
(0.110)
1.101***
(0.152)
0.256***
(0.0594)
0.181**
(0.0849)
0.308**
(0.146)
0.180**
(0.0702)
0.219**
(0.0901)
-0.0538
(0.158)
0.212***
(0.0800)
-0,548*
(0,255)
-0.548**
(0.0594)
-0.610***
(0.231)
-0,643***
(0,0755)
-0.415
(0,278)
-0.463***
(0.116)
7.28e-09**
(3.36e-09)
-8.21 e-08*
(4.28e-08)
-1.34e-07*
(8.00e-08)
1.199-07***
(3.32e-08)
8.74e-09**
(3.88e-09)
-1.52e-07**
(6.78e-08)
2.33e-07***
(7.75e-08)
7.03e-08
(1.27e-07)
6.04e-09*
(3.41 e-09)
-9.73e-08**
(4.32e-08)
-5.06e-08
(7.96e-08)
9.05e-09**
(3.86e-09)
-1.58e-07**
(6.23e-08)
2.43e-07***
(7.78e-08)
Businesses
-4.60e-07
Wage
(4.80e-07)
8.31e-05***
(2.16e-05)
Time 85-87
Michigan
Treatment in
Michigan
Income
Dividend
Rent
Income
Population
8.69e-08***
6.28e-08
(3.27e-08)
(1.26e-07)
-1.23e-06**
-6.1Oe-08
-1.21e-06**
(5.48e-07)
0.000173***
(3.88e-05)
(4.89e-07)
7.37e-05***
(2.14e-05)
(5.52e-07)
0.000177***
(3.87e-05)
-0.00875**
(0.00432)
-0.0210***
(0.00809)
-0.00440
(0.00432)
-0.0218***
(0.00805)
500
490
500
490
House
prices
Observation
s
500
490
Number of
49
Robust standard errors in parentheses *** p<0.01, * p<0.05, * p<0.1 1
stid
69 Asset Sharing and Stakeholder Arrangements
49
49
Table 4 (continued): Poisson Pooled and Fixed Effect Regressions
(3)
(2)
(1)
Models
Pooled Poisson
Fixed Effect
Poisson
% Technical
%4Yr
College
Union
Unemployed
Covenants
(5)
Pooled Poisson
With Control
Fixed Effect
Poisson with
Controls
Pooled Poisson
with Controls
and Regional
2.045
(1.811)
-1.891
(2.500)
0.765
(1.806)
-2.047
(2.562)
-4.809*
(2.610)
-0.0159*
(0.00924)
0.0116
(0.0212)
-1.313
(3.341)
-0.0180
(0.0293)
0.000417
(0.0243)
-1.684
(2.611)
-0.0166*
(0.00944)
0.0116
(0.0209)
-1.461
(3.282)
-0.0160
(0.0294)
-0.00135
(0.0247)
.0374
0,149**
0.172***
(.0409)
(0.0330)
(0-0331)
0.796***
(0.145)
Regional
Regional
-0.199
(0.179)
Treatment
Constant
0.838***
-1.434**
-1.974***
(0.0652)
(0.717)
(0.696)
Observation
s
(6)
Fixed Effect
Poisson with
Controls and
Regional
Controls
(4)
500
Number of
stid
490
500
49
Robust standard errors in parentheses
*** p<0.01, **p<0.05, * p<0.1 1
70 1Asset Sharing and Stakeholder Arrangements
490
49
500
-0.218*
(0.133)
490
49
Table 5: Poisson Pooled and Fixed Effect Regressions - Incident Ratio
Models
(4)
(5)
(6)
(1)
(2)
(3)
Pooled Poisson
Fixed Effect Poisson
Pooled Poisson
With Control
1.292**
1.292***
1.198**
1.197**
1.245**
1.237***
(0.142)
(0.0768)
(0.102)
(0.0840)
(0.112)
(0.0989)
Fixed Effect Poisson
Pooled Poisson
Fixed Effect Poisson with Controls and with Controls and
with Controls
Regional Controls Regional Controls
Time 85-87
Michigan
3.009***
1.361 **
0.948
(0.458)
(0.199)
(0.150)
Treatment in
Michigan
0.578**
(0. 147)
o.578***
(0.0343)
0.544***
0.526***
0.660
0.630***
(0.125)
(0.0397)
(0.184)
(0.0729)
1.000**
1.000**
1.000*
1.000**
Income
(3.36e-09)
(3.88e-09)
(3.41 e-09)
(3.86e-09)
1.000*
1.000**
1.000**
1.000**
(4.28e-08)
(6.78e-08)
(4.32e-08)
(6.23e-08)
Dividend
Rent Income
1.000*
1.000***
1.000
1.000***
(8.00e-08)
(7.75e-08)
(7.96e-08)
(7.78e-08)
1.000***
1.000
1.000***
1.000
Population
(3.32e-08)
(1.27e-07)
(3.27e-08)
(1.26e-07)
1.000
1.000**
1.000
1.000**
(4.80e-07)
(5.48e-07)
(4.89e-07)
(5.52e-07)
Businesses
Wage
1.000***
1.000***
1.000***
1.000***
(2.16e-05)
(3.88e-05)
(2.14e-05)
(3.87e-05)
0.991**
0.979***
0.996
0.978***
(0.00428)
(0.00792)
(0.00430)
(0.00788)
House Prices
Constant
1.967***
0.238**
0.139***
(0.390)
(0.171)
(0.0967)
Observations
500
490
500
490
Number of stid
49
Robust standard error in parentheses
*** p<0.01, **p<0.05, * p<0.1I
711 Asset Sharing and Stakeholder Arrangements
49
500
490
49
Table 5(continued): Poisson Pooled and Fixed Effect Regressions - Incident Ratio
Models
(2)
(3)
Fixed Effect Poisson
Pooled Poisson
With Control
(1)
Pooled Poisson
(4)
(5)
(6)
Pooled Poisson Fixed Effect Poisson
Fixed Effect Poisson with Controls and with Controls and
with Controls
Regional Controls Regional Controls
% Professional
1.928
0.286
70.34*
0.411
(4.060)
(1.436)
(154.1)
(2.076)
% Educated (+6
yrs)
166.7*
1.287
214.6*
1.844
(492.3)
(3.885)
(606.3)
(5.779)
7.731
0.151
2.150
0.129
(14.00)
(0.377)
(3.883)
(0.331)
%Technical
%4 Yr College
Union
0.00815*
0.269
0.186
0.232
(0.0213)
(0.899)
(0.485)
(0.761)
0.984*
0.982
0.983*
0.984
(0.00910)
(0.0288)
(0.00928)
(0.0290)
Unemployed
Covenants
1.012
1.000
1.012
0.999
(0.0215)
(0.0243)
(0.0212)
(0.0247)
1.038
1,160**
1.188***
(.04252)
(0.0383)
(0.0393)
Regional
2.216***
(0.322)
Regional Treatment
Constant
0.820
0.804*
(0.147)
(0.107)
1.967***
0.238**
0.139***
(0.390)
(0.171)
(0.0967)
Observations
500
490
500
Number of stid
Robust standard error in parentheses
49
*** p<0.01, **p<0.05, * p<0. 1
72 1Asset Sharing and Stakeholder Arrangements
490
49
500
490
49
Table 6: Michigan Yearly Effects Pooled and Fixed Effect Poisson - Incident Ratio Rates
(2)
(1)
Models
Fixed Effect Poisson
Pooled Poisson
Michigan
1983
1983 Michigan Treatment
1984
1984 Michigan Treatment
1985
1985 Michigan Treatment
1986
1986 Michigan Treatment
1987
1987 Michigan Treatment
1988
1988 Michigan Treatment
1989
1989 Michigan Treatment
1.197
(0.237)
1.188
(0.191)
1.401
(0.300)
1.520***
(0.207)
1.204
(0.242)
1.896***
(0.238)
0.587***
(0.114)
1.389**
(0.190)
0.891
(0.178)
1.311
(0.225)
0.396***
(0.0823)
1.506***
(0.217)
1.270
(0.259)
1.428**
(0.240)
0.973
(0.210)
Includes All Controls
Robust standard errors in parentheses
*** p<0.01, **p<0.05, * p<O.l
73 Asset Sharing and Stakeholder Arrangements
1.437**
(0.215)
1.392***
(0.165)
1 .688***
(0.194)
1.080
(0.118)
1.982***
(0.208)
0.543***
(0.0601)
1.479***
(0.155)
0.775**
(0.0861)
1.373***
(0.167)
0.367***
(0.0390)
1.447***
(0.164)
1.086
(0.127)
1.259**
(0.131)
0.873
(0.109)
74 1Asset Sharing and Stakeholder Arrangements