|
|
|
VII. THE RELATIONSHIP BETWEEN FINANCE, CORPORATE GOVERNANCE, AND
LABOR
Mandated Codetermination and Firm Performance: The Productivity
Effects of German Works Councils Revisited
Bernd Frick
University
of Witten/Herdecke
Introduction The recent political debate about the latest reform of the
German "Works Constitution Act" has led to an unprecedented increase in
the public interest regarding the relative impact of alternative forms
of worker participation on firm performance. Moreover, the academic discussion
by labor and personnel economists has been spurred by a number of different,
though closely related developments:
- The increasing globalization of product and labor markets
has led to a resurgence in academic interests regarding the German system
of industrial relations. On the one hand it has been argued that the German
system may have to surrender to the pressures of competition. On the other
hand, it has been argued that specific German idiosyncrasies (like the
influence of large banks and the existence of mandated works councils)
may be a source of competitive advantage that is unlikely to be eradicated
by globalization. Although most of the arguments discussed in this context
are well known, it is interesting to see how the formerly incompatible
theoretical positions seem to have converged recently.
- The availability of several large and representative firm panels has fostered
empirical analyses that have been impossible to conduct before. These
data sets have been widely used to document the influence of mandated
works councils on various dimensions of firm performance. However, the
main finding of most of the available research--that, other things equal,
the existence of a works council has no pronounced influence on the performance
of firms (positive effects on labor productivity and personnel turnover
are compensated by a negative influence on profits)--is problematic for
at least one reason: most authors concentrate their attention on investments
in physical capital, thereby neglecting investments in human and organizational
capital which, in turn, may be of paramount importance for the works councils.
Thus, the findings presented so far may lead to inadequate policy implications.
The following contribution to the growing body of literature
has three different goals. First, I want to review the theoretical arguments
that have been raised in the more recent discussion. In this context I
show that the formerly incompatible positions have converged to a considerable
degree. Second, I look at the impact of works councils on firms' investments
in human and organizational capital and then present different production
function estimates that have been augmented by, inter alia, a works council
dummy. I conclude with a summary of the main findings and some implications
for further research.
Works Councils: Welfare Reducing Cartels or Efficiency
Enhancing Institutions? Assuming that ownership accompanied by secure property rights
is the most effective institution for providing individuals with incentives
to create, maintain, and improve assets, it is argued that it is also
essential that the residual rights of control, i.e., the rights to make
any decisions concerning an asset's use, are exclusively controlled by
a single party. The economic importance of residual control follows from
the difficulty of writing contracts that specify all the control rights.
This, however, implies that the parties to a contract are able to foresee
all future developments and can write down and enforce a complete contract--one
that specifies what each party has to do in every relevant eventuality
at every future date and how the resulting income in each such event should
be divided. However, complete contracts are generally impossible for transactions
of any significant complexity that occur over a period of more than just
a few days (Milgrom and Roberts 1992, 288-89). Consequently, arrangements
that leave all control rights--that are not otherwise assigned--to a single
party (eliminating the need to negotiate and reach agreement for every
unanticipated development) may result in significant cost advantages.
While the notion of ownership as residual control is relatively clear
for a simple asset, it gets increasingly fuzzy when applied to a (large)
firm. Decisions by the owner or the management may be especially controversial
when not only the physical capital of the firm, but also the human capital
of the firm (employees) is affected.
Jensen and Meckling (1979), for example, suggest that when the party having
residual control rights is also entitled to receive the residual returns,
then the residual decisions made tend to be efficient ones. More specifically,
they argue that in a firm, where the workers receive contractually agreed
upon fixed wages in exchange for the effort they supply, the residual
claimant will, just by pursuing his own interests and maximizing his returns,
make efficient decisions. Under these assumptions, a redistribution of
control rights will necessarily lead to an inefficient resource allocation,
because those who bear the residual risks are not the only party to decide
on the use of the firm's assets. These arguments, in turn, form the basis
of their market-oriented case against mandated codetermination published
more than twenty years ago: "If codetermination is beneficial to both
stockholders and labor, why do we need laws which force firms to engage
in it? Surely, they would do so voluntarily. The fact that stockholders
must be forced by law to accept codetermination is the best evidence we
have that they are adversely affected by it" (Jensen and Meckling 1979,
474).
More recently, this orthodox position has been challenged by a number
of economists, be they proponents or critics of property rights theory.
First, it has been argued that decisions made by the residual claimant
may not always be efficient: if only part of the costs of a decision accrue
to the party making the decision, then that party will find it in its
interest to ignore some of the external effects, sometimes leading to
inefficient decisions. If, for example, efficient production requires
that workers invest in firm-specific skills, then institutions that protect
their investments make them more likely to invest in acquiring those skills.
According to Furubotn (1988, 167), "Workers who undertake durable reliance
investments commit themselves to the firm for some time into the future
and are, therefore, vulnerable. The distribution of the firm's quasi-rents
and the value of the labor assets can be affected by the behavior of other
members of the coalition. Hence, the possibility exists that worker-investors,
if unprotected by institutional or contractual safeguards, may be exploited
and suffer serious economic injury." Thus, if workers are not protected
by institutional or contractual safeguards against the opportunistic behavior
of other members of the coalition, they will either be unwilling to invest
in the acquisition of firm specific skills or may risk serious economic
loss in the case of dismissal. In a situation, where not all of the coalition-specific
resources are owned by a single party, codetermination is likely to be
a governance structure that is capable of dealing with maximizing agents
with conflicting interests. However, irrespective of this generally favorable
view of voluntary codetermination, legal intervention by the state is
unequivocally rejected (Furubotn 1988, 178).
Second, the Jensen-Meckling argument has recently been challenged by,
among others, Freeman and Lazear (1995), who argue that codetermination
is likely to be underprovided by the market. Cooperative solutions to
the prisoner's dilemma are assumed not to occur as long as there is no
exogenous regulation by some third party. However, although mandated works
councils have the potential to foster an increase in the joint surplus,
firms are most likely to oppose them, according to Freeman and Lazear
(1995, 29), because "institutions that give workers power in enterprises
affect the distribution as well as amount of joint surplus. The greater
the power of works councils, the greater will be workers' share of the
economic rent. If councils increase the rent going to workers by more
than they increase total rent, firms will oppose them. It is better to
have a quarter slice of a 12-inch pie than an eighth slice of a 16-inch
pie." Given these seemingly incompatible positions, theory offers no definitive
guidance as to the likely effects of mandated codetermination. The beneficial
and detrimental effects must be demonstrated empirically. The following
section discusses some recent evidence.
Codetermination, Organizational Capital, and Economic Performance Until recently, the number of studies analyzing the influence
of works councils on firm performance was rather low and their quality
tended to be poor. With the availability of different firm panels, the
situation has changed quite dramatically; the number of studies has been--and
still is--increasing rapidly, and the more recent studies suffer less
from methodological problems than the ones that had been published up
to the mid- and late 1990s.1 The more recent studies have used a variety
of measures, including productivity levels and growth, financial performance
and profitability, investment in research and development, and job generation.
Summarizing these studies, it appears that works councils seem to have
no clear cut consequences for firm performance. On the one hand, the presence
of a works council has--other things being equal--a significantly positive
influence on labor productivity, but a significantly negative influence
on labor costs and profitability. On the other hand, works councils seem
to have no influence on investment behavior or on innovations (neither
on product nor on process innovations) (Addison et al. 2001; Jirjahn 2003;
and Dilger 2002).
These findings are neither compatible with the notion of works councils
as "rent-seekers" that tend to ignore the interests of owners and managers
nor with the argument that firms would benefit from the existence of a
legally mandated plant-level representation. Given these results, it is
worth extending the analysis of the impact of works councils to some "intangible
assets" that--although of paramount importance for the short- and the
medium-term performance of the firm--are very often neglected by economists:
the credibility of long-term career promises, the readiness to finance
and to participate in further training, and the acceptance of organizational
change.
Works Councils and Personnel Turnover In order to maximize worker effort, loyalty, and motivation,
firms usually implement specific incentive mechanisms to avoid opportunistic
behavior. Since the deposition of bonds or "entrance fees" is neither
feasible nor legally enforceable, workers are initially paid less than
their marginal product, but eventually are paid a wage exceeding their
marginal product. Over the expected tenure with the firm, workers receive
an expected present value of compensation equal to the present value of
their productivity (Lazear 1981). However, once a worker has posted a
bond in the form of a wage below his productivity, the firm has a strong
incentive to label him a "shirker" and to claim his bond. When workers
are uncertain of the trustworthiness of firms, they are unlikely to be
willing to post such bonds. In this context, a works council may serve
as a credible institution that can be relied upon by both parties to determine
whether a worker has shirked.
Using different representative samples of large numbers of private sector
firms in (East and West) Germany, Frick (1996), Frick and Möller
(2003), Dilger (2002) and Addison, Schnabel and Wagner (2001) find that
firms with a works council have significantly lower dismissal and quit
rates than observationally similar firms without plant-level interest
representation. Additional estimates reveal that works councils neither
oppose dismissals in shrinking firms nor inhibit hirings in growing firms.
Moreover, works councils seem to exert no influence on the structure of
the dismissed worker population, suggesting that the interests of the
firm are usually taken into account by the workers' representatives (who
are often said to unequivocally favor the interests of older workers with
long tenure and poorer opportunities in the external labor market).
Works Councils and Training Expenditures Depending on the transferability of the acquired skills, training
increases a worker's productivity roughly to the same extent in his current
firm only or in a number of alternative employment relationships. Depending
on the type of human capital acquired, the party that has to amortize
the initial training costs (or the larger share thereof) may incur serious
economic losses, which, if anticipated would cause the party not to invest
at all (Gerlach and Jirjahn 2001). Thus, a hold-up problem arises when
one party can ex post expropriate parts of the surplus of a specific investment
undertaken by the other party, thus causing disincentives for investing
in specific human capital. Moreover, there exists a serious poaching problem;
irrespective of the transferability of general skills, many firms invest
in the training of their workers. This behavior is due to information
asymmetries resulting from the deficiencies of imperfect labor markets.
In this case, firms may withhold investments because they fear that other
firms may try to hire their trained workers without incurring any training
costs themselves.
Irrespective of the specific sharing rule according to which the training
costs have to be borne by employer and employee, a works council can monitor
both parties' potentially opportunistic behavior. Moreover, the poaching
problem can be mitigated. Using a representative panel of firms located
in Lower Saxony, Gerlach and Jirjahn (2001) show that the presence of
a works council has a significant positive influence on the probability
that firms invest in the training of their respective workforces as well
as on the training expenditures per employee. Thus, the positive impact
of works councils on further training is in accordance with the hypothesis
that mandated codetermination promotes cooperative and trustful industrial
relations which alleviate many of the market failure problems resulting
from employer-provided further training.
Works Councils and Organizational Flexibility Using data from a representative longitudinal survey of engineering
firms, Dilger (2002) analyzes the influence of works councils on working-time
arrangements. The data used in this study is novel insofar as it not only
allows a distinction between firms that have a works council and those
that do not, but also permits a closer characterization of the works council
as viewed by the respective firm's management. Dilger finds that only
two out of five "types" of works councils are beneficial to the firm with
regard to the introduction of flexible working-time arrangements. While
works councils that are considered to be a "tough" or a "cooperative"
partner seem to foster such arrangements, this is not the case when they
are "antagonistic," "disinterested," or when they are not involved by
management in the decision-making process. Thus, although works councils
may be seen as an institution that reduces the firm's external flexibility
(however, the empirical findings quoted above do not seem to support that
assumption), they apparently increase internal flexibility by promoting
the introduction of working time arrangements that deviate from "standard"
working hours among their constituents.
Moreover, a study by Frick (2002)--also using the above mentioned panel
from the machine tool industry--finds that it is not the presence of a
works council per se that influences the adoption of high performance
work practices, but its "level of activities"--measured by the number
of firm-level agreements concluded during the last three years--and the
"type" of the works council as viewed by the management of the firm. In
firms with an "antagonistic" works council, the number of high performance
work practices is higher than in otherwise similar firms that have either
a "disinterested" or even an "excluded" works council. Looking at the
performance effects of such practices, it becomes apparent why the works
councils often reject their introduction: other things being equal, the
adoption of these practices increases expected as well as actually realized
firm performance (measured by changes in product demand, in sales, and
in profitability), but at the same time they reduce the demand for labor.
This means that firms do indeed benefit from such practices--but very
often at the expense of their workers.
Works Councils and Firm Performance The fact that works councils can--under specific circumstances--act
as an institution to reduce the probability of opportunistic behavior,
is only a necessary but not a sufficient condition to document their efficiency.
In order to be able to refute the assumption that works councils lead
to an inefficient allocation of resources, it is indispensable to document,
inter alia, a positive influence of mandated workers' representation on
labor productivity. Due to data limitations, very few of the available
studies have been able to control for the capital stock of the companies
in the samples used and were, therefore, unable to rule out the possibility
that it is capital intensity rather than the presence of a works council
that fosters the economic performance. One of the first papers to overcome
this methodological problem is Frick and Möller (2003) who used two
different waves (1998 and 2000) from the largest firm panel currently
available in Germany, the IAB Panel, with information from some four thousand
firms in East as well as in West Germany. Assuming that financial means
that have recently been invested to replace used capital goods are highly
correlated with the capital stock, the authors estimate different types
of production functions (Cobb-Douglas, CES, and Translog) with value added
as the dependent variable. Apart from information on capital and workers
employed, the production function estimates include a wide range of variables
identified as (potential) determinants of firm performance, such as the
percentage of women and part-time workers, product and process innovations,
exports, training intensity, product market competition, quality of technical
equipment, R&D expenditures, etc.
The estimates reveal that--other things being equal--the presence of a
works council has a positive and statistically significant influence on
the economic performance of German firms (measured by value added). The
respective coefficients of the works council dummy indicate that these
effects are rather large and that they differ significantly between industry
and service sectors. The values of the coefficients obtained indicate
that in 1998 in West German firms the presence of a works council increases
labor productivity by about 25 percent, while in East German firms the
respective figure is about 30 percent. Repeating the estimates with data
from the year 2000 confirms the initial findings. It is worth mentioning,
however, that the positive impact of works councils on labor productivity
seems to be much more pronounced in the service sector. Distinguishing
between manufacturing and service firms, it appears that the point estimate
for service firms in East Germany is twice the size of that for manufacturing
firms. For West German firms, the works council dummy is even insignificant
for manufacturing firms, but highly significant for service firms. Moreover,
collective bargaining coverage combined with the presence of a works councils
has no impact in industry, but leads to a lower labor productivity in
the service industry in both parts of the country.
Summary and Implications Based on "modern" concepts of the firm (emphasizing the importance
of bounded rationality and opportunistic behavior), the perception of
mandated workers' representation has changed considerably during the 1990s.
To the extent that a credible works council can convince the firm's workforce
to accept the implementation of measures that seem to violate their expectations,
mandated codetermination is likely to overcome the problems inherent in
a "prisoner's dilemma" situation, where credible commitments are impossible
to be made without the support of an exogenously implemented institution,
whose task is to monitor the behavior of the contracting parties.
The most important empirical findings can be summarized as follows. First,
a review of the literature on the influence of works councils on investments
in intangible assets (such as personnel turnover, organizational change,
and further training) suggests that concentrating on investments in physical
capital, on productivity, profitability, investments, and some other easy
to measure indicators of firm performance may lead to a considerable underestimation
of the positive effects of mandated works councils. Second, the presence
of a works council has a positive and statistically significant influence
on the economic performance of German firms. The respective coefficients
indicate that these effects are rather large and that they differ significantly
between industry and service sectors.
The major shortcomings of the available evidence are obvious. We do not
yet know whether the productivity increase induced by mandated works councils
is large enough to compensate firms for the associated increase in labor
costs. Moreover, we cannot yet reject the hypothesis that the productivity
increases associated with mandated works councils may be the result of
some omitted variables, such as an especially competent management. However,
the size and the significance of the works council dummy in a number of
studies using different data sets and different model specifications make
this argument less convincing. Moreover, given the plausibility of the
theoretical arguments suggesting a positive influence of works councils
on firm productivity, there are few reasons to expect that including measures
of management quality would yield results significantly different from
the ones summarized above.
Notes
1. For an overview of recent
studies see Addison et al. (2001). Studies that have been published prior
to 1996 are summarized by Frick and Sadowski (1995).
References
Addison, John T., Claus Schnabel, and Joachim Wagner. 2001.
"Works Councils in Germany: Their Effect on Establishment Performance."
Oxford Economic Papers, 53, pp. 659-94.
Dilger, Alexander. 2002. "Ökonomik betrieblicher Mitbestimmung: Die
wirtschaftlichen Folgen von Betriebsräten." Münich: Hampp.
Freeman, Richard B., and Edward P. Lazear. 1995. "An Economic Analysis of
Works Councils." In Works Councils: Consultation, Representation, and
Cooperation in Industrial Relations, ed. Joel Rogers and Wolfgang Streeck.
Chicago: University of Chicago Press, pp. 27-50.
Frick, Bernd. 1996. "Co-determination and Personnel Turnover: The German
Experience." Labour, Vol. 10, pp. 407-30.
Frick, Bernd. 2002. "High Performance Work Practices und betriebliche Mitbestimmung:
Komplementär oder substitutiv?" Industrielle Beziehungen, Vol.
9, pp. 79-102.
Frick, Bernd, and Iris Möller. 2003. "Mandated Works Councils and Firm
Performance: Labor Productivity and Personnel Turnover in German Firms."
Journal of Applied Social Science Studies, Vol. 123, pp. 423-54.
Frick, Bernd, and Dieter Sadowski. 1995. "Works Councils, Unions, and Firm
Performance: The Impact of Workers' Participation in Germany." In Institutional
Frameworks and Labor Market Performance, ed. Friedrich Buttler, Wolfgang
Franz, Ronald Schettkat, and David Soskice. London: Routledge, pp. 46-81.
Furubotn, Eirik G. 1988. "Codetermination and the Modern Theory of the Firm:
A Property Rights Analysis." Journal of Business, Vol. 61, pp. 165-81.
Gerlach, Knut, and Uwe Jirjahn. 2001. "Employer Provided Further Training:
Evidence from German Establishment Data." Journal of Applied Social Science
Studies, Vol. 121, pp. 139-64.
Jensen, Michael C., and William H. Meckling. 1979. "Rights and Production
Functions: An Application to Labor-Managed Firms and Codetermination." Journal
of Business, Vol. 52, pp. 469-506.
Jirjahn, Uwe. 2003. "Produktivitätswirkungen betrieblicher Mitbestimmung--welchen
Einfluss haben Betriebsgröe und Tarifbindung?" Zeitschrift
für Betriebswirtschaft, Supplement no. 4, Vol. 73, pp. 63-85.
Lazear, Edward P. 1981. "Agency, Earnings Profiles, Productivity, and Hours
Restrictions." American Economic Review, Vol. 71, pp. 606-20.
Milgrom, Paul, and John Roberts. 1992. Economics, Organization and Management.
Englewood Cliffs, N.J.: Prentice-Hall.
.
|
