Digitized by the Internet Archive
in
2011 with funding from
Boston Library Consortium IVIember Libraries
http://www.archive.org/details/privatizationmanOOschm
l^^iY
HB31
.M415
working paper
department
of economics
PRIVATIZATION AND MANAGEMENT INCENTIVES
IN THE TRANSITION PERIOD IN EASTERN EUROPE
Klaus M. Schmidt
Monika Schnitzer
No.
92-17
Nov.
1992
massachusetts
institute of
technology
50 memorial drive
Cambridge, mass. 02139
M.I.T.
LI^IAR^
PRIVATIZATION AND MANAGEMENT INCENTIVES
IN THE TRANSITION PERIOD IN EASTERN EUROPE
Klaus M. Schmidt
Monika Schnitzer
No.
92-17
Nov.
1992
^AR_3
1
1993
I
Privatization and
in the Transition
Management
Incentives
Period in Eastern Europe*
Klaus M. Schmidt
Massachusetts Institute of Technology and Bonn University
Monika Schnitzer
Massachusetts Institute of Technology and Bonn University
First version:
June 1992
This version: November 1992
Abstract: The paper
develops a simple theoretical framework in which
the impact of different governance structures on
management
incentives, the
efficiency of restructuring, and the social costs of the adjustment process in the
transition period in Eastern
Europe can be analyzed.
The model shows
that
immediate privatization leads to strong management incentives to restructure
but also to high social costs of bankruptcies and layoffs. If the government
stays in control social costs will be lower. However, in this case
"soft
budget constraint" and have
less incentives to restructure.
suggests which companies should be privatized
Keywords:
Privatization,
managers face a
The model
also
first.
Economic Reform
in
Eastern Europe, Principal-
Agent Theory.
*This paper was prepared for the "Conference on the Structure and Behavior of Economic Organizations:
Comparative Theoretical and Empirical Perspectives", held at the University of Minnesota September 1-3,
1992. We would like to thank Avner Ben-Ner, Olivier Blanchard, Josef Brada, Stanley Fischer, Oliver Hart,
Gustav Horn, and Egon Neuberger for helpful comments and discussions.
Introduction
1.
Privatization
is
probably the most important and the most
element in the trans-
difficult
formation process of the economies in Eastern Europe. Although considerable progress has
been made in privatizing small and medium sized firms, the bulk of large industrial enterprises
is still
owned and
owned by the
state.
Companies
comparable
controlled within a complex system of private
developed over decades to their present form.
over night.
of
The question
in the transition period,
is
It is
size in
and public
western economies are
institutions
clearly impossible to create such a
what governance structure should be established
i.e.
before a
which have
modern market economy with
system
for these firms
all its institutions
has
been developed.
Several economic advisers and academic economists have offered detailed proposals
how
and/or privatize these firms. Their arguments in favor or against different
to restructure
schemes are based on the experiences made with privatization in other countries and the
basic lessons of economic theory.
We
formation process.
feel,
These arguments are most valuable to guide the trans-
however, that at this stage some additional insights can be
obtained by making these ideas more precise in formal models designed to capture the main
peculiarities of the transition phase.
retical
framework
structures on
in
The purpose
of this paper
is
to develop a simple theo-
which we can analyze systematically the impact of different governance
management
incentives, the efficiency of restructuring,
and the
social costs of
the adjustment process.
Our
analysis focuses
of output
and employment and pose the greatest challenge
overstaffed
it is
on large industrial companies. They account
and need a
lot of painful restructuring before
difficult to find private investors for
for
for reform:
they
a high proportion
many
may become
of
them
profitable,
are
and
them.
We distinguish two idealtypical approaches
which lead to two different governance struc-
tures for these companies in the transition period. Proponents of the market approach argue
that the best
large state
way
to proceed
owned companies
lack of entrepreneurs
is
in
to privatize as fast as possible
by distributing the shares of
one way or another to the general population.
Given the
and the atomistic ownership structure resulting from a mass distribu-
1
financial intermediaries (like holding companies,
scheme
tion
come
or investment banks) will
diversify risk
in
and play a
They
crucial role in the transition period.
will
and actively control the managers of the companies they own.
Proponents of the government approach suggest that at
who commits
individually to a core investor
To
several years.
an entrepreneur
find such
which the company
is
owned by the
some coarse restructuring
may
his investment to the
each
company
amount
take a considerable
sell
more
detail in Section 2.
for
of time in
and controlled by the government agency.
state
describe these idealtypical approaches in
least
Thereafter this agency should try to
should be done by a government agency.
company
mutual funds, pension funds,
We
-
Before we model different governance structures we have to specify the economic envi-
ronment
which they are supposed to work. Our working hypothesis
in
is
that the transition
period in Eastern Europe will be characterized by rapidly changing market conditions, eco-
nomic and
to well
political instability,
known
and a high degree
results in contract theory
we
be used in such an environment in Section
tions
of uncertainty
what kind
discuss
3.
Among
and performance based compensation schemes
period, but that the threat of bankruptcy
may
about the future. Referring
of incentive contracts should
other things
will
we argue
that stock op-
be of limited use in the transition
play an important role to motivate managers
to restructure their firms efficiently.
In Section 4
we introduce the formal model which captures the two
nance structures of the market and the government approach and which
discussion of feasible incentive schemes in the transition period.
effects:
First,
it
shows that
if
reflects the informal
The model generates two
firms are controlled by a government agency then the social
costs of the adjustment process will be lower than
in control.
idealtypical gover-
The government agency
if
will take profits
a privately
owned holding company
is
from successful firms to cross-subsidize
unsuccessful ones. This reduces the social costs of liquidation, but the government agency
may
subsidize inefficiently
much and keep
firms in operation which should be closed down.
This delays the necessary adjustment process.
The second
efficiently
if
his
ernment agency.
effect
is
company
that the
is
manager
owned by a
The reason
is
of a firm will
work harder and restructure more
private holding than
that the
if it is
manager knows that
controlled by the gov-
his firm
is less
likely to
go
^
bankrupt under government control.
If his
firm
fails
the government will try to rescue the
Thus the manager
firm in order to prevent the social costs of liquidation.
budget constraint": He
enough
In our
through a painful restructuring of his firm or spend
modernize the production process because he rationally anticipates that the
effort to
government
will not carry
faces a "soft
will bail
him out
in case
model privatization
unsuccessful firms.
A
he
fails.
a commitment device of the government not to subsidize
is
made by Sappington and
similar point has been
Stiglitz (1987)
who
argue that public ownership generally facilitates intervention by the government, while intervention
more
is
difficult if
the enterprise has been privatized. Intervention
allows rapid adaptation to unforseen contingencies, but
the
management
We
try to
make
to control costs, because the
this
argument more precise
also
it
government
in a formal
is
beneficial
if it
undermines the incentives of
is likely
to subsidize cost overruns.
model which captures some of the
distinctive features of the transition period in Eastern Europe.
In Section 5
we
point out
some implications
of the
model and discuss intermediate
governance structures. For example, the government need not privatize
same time. Our model
gives
some
insights into
all
the firms at the
which firms should be privatized immediately
and which ones should be kept under the control of a government agency during the transition
period. Also, the
model shows that
it
may be
fraction of the shares which yields a limited
firms too much.
A
similar
beneficial for the
commitment not
commitment can be obtained
if
government to keep only a
to cross-subsidize unsuccessful
the government owns no shares
but assumes the right to replace the directors of a privately owned holding company. Section
6 concludes.
2.
The
first
Privatization Proposals
step of any .privatization
In the last years of socialism
and
program
is
the establishment of clear ownership
in the early transition period the
rights.
governments of most
Eastern European countries gave up important ownership and control rights which are now
'This effect
Kornai (1980).
is
similar to the "soft budget constraint" for
managers
in socialist
economies as discribed by
exercised de facto and often de jure by workers' councils, managers and sometimes local
authorities.
The
current situation
is
muddled ownership structures
characterized by
the control rights of the different stakeholder groups are
ill
companies as a compensation
Most economic advisers
tized" (or "commercialized"),
and
teries
up
set
them considerable
it is
Most privatization
fractions of the stock of their
for the loss of control.
companies should be "corpora-
also agree that large industrial
i.e.
they should be formally separated from the branch minis-
as legally independent corporations."^
such a corporation
which
defined and in conflict with each
other. ^ Privatization can only go ahead with the consent of these groups.
proposals recognize this fact by offering
in
no longer involved
in the
Although the government
day to day
activities of the firm
still
owns
but rather
and the performance of the management through a board of
controls the strategic decisions
directors. Corporatization facilitates the second step,
companies to private investors.
How
this should
which
to transfer ownership of these
is
be done, however,
is
very controversial.
Goals of privatization
Before
we
discuss alternative privatization proposals let us briefly
to be accomplished through corporatization
-
Efficient
management
managers the
-
control:
in
the transition period:
Governance structures have to be established which give
right incentives to run
Restructuring of the economy:
and privatization
summarize the main goals
The
and restructure
their
companies
large industrial firms are
more
zontally integrated than comparable firms in western economies.
are
still
hoarding too
much
workforce. Furthermore,
to be closed
efficiently.
vertically
Many
and
hori-
of these firms
labor and will have to lay off a significant fraction of their
some companies
are expected to be non viable at
down, either because the markets
for their
all
and have
products have disappeared, or
because their production technologies are hopelessly outdated.
-
Limit the social, costs vf adjustment:. Even, if .a laxge-company
profitable in the future,
^This problem
is
it
may
particularly severe in Poland,
(1990) and Shieifer and Vishny (1992).
^See
e.g.
be socially
Fischer (1991).
inefficient to shut
it
is
not expected to be
down immediately. High
Hungary and Russia. See Dabrowski
et.al.
(1991), Kornai
unemployment
rates
and the economic desolation of entire regions may have consider-
able negative external effects which have to be taken into account. Furthermore, social
and
-
political unrest
may jeopardize
the transition process as a whole.
Attract foreign capital and expertise: Given the lack of domestic savings foreign invest-
ments
an important role for economic growth. Furthermore, foreign invest-
will play
ments are expected to lead to a transfer of technological knowledge and managerial
techniques which are crucial to modernize the economy.
-
-
Create competitive market structures:
Some conglomerates have to be
split
up to reduce
their
market power, and the government has to regulate natural monopolies.
Fair
initial distribution of wealth:
state,
it is
widely
felt
Since the existing capital stock
is
owned by the
that each adult citizen has an equal claim on these assets, and
that capitalism should start with an egalitarian distribution of this wealth.
-
Protect governm,ent revenues: In
mainly through the transfer of
all socialist
countries government revenues were raised
profits to the state budget, while a tax
parable to that in western industrialized countries did not exist.
If all
system com-
owned
state
companies were privatized before a functioning tax system has been established, the
state
would
lose its
main source
of revenues.
Naturally, different privatization schemes put different weights on different objectives.
ever, the
main
distinctive feature of the different proposals
achieving these objectives during the transition period.
is
Two
How-
the role of the government in
idealtypical approaches can be
distinguished. Proponents of the "market approach" believe that 40 years of socialism have
shown that governments are incapable
Thus, the best way to proceed
market
forces,
with as
little
is
of guiding the efficient restructuring of the
to privatize as rapidly as possible
government interference
economy.
and to leave the
rest to
as possible. In contrast to this position
proponents of the "government approach" stress the lack of entrepreneurs and of established
market institutions
in Eastern
Europe. There
is
no well functioning stock market and no
system of experienced financial intermediaries to rely on. They claim that, given the scope
of the transformation
and the
risks involved, a strong lead
by the government
is
unavoidable.
In the following
ever,
this
we
will outline these
two idealtypical approaches
most actual privatization schemes take an
eclectic position.
in
It is
more
detail.
beyond the scope
paper to discuss this literature and the recent developments in Eastern Europe in
For our purpose
it is
sufficient to
Howof
detail.
overview the basic arguments.'*
The market approach:
Ownership and control
rights should
be transfered to private investors as
fast as possible.
Since privatization through initial public offerings or individual sales would take very long
(see the
arguments given below), a mass privatization scheme with
has to be employed.
A
minority fraction of (non-voting) shares could be given to workers
and managers to compensate them
some
free distribution of shares
for their loss of control.
The government could
also retain
fraction of shares as a source of future revenues, but this fraction should be either non-
voting or sufficiently small that the government cannot interfere with the strategic decisions
of the companies.
The majority
of shares will then be distributed either directly
through a voucher scheme
to the general public, or indirectly to financial intermediaries like holding companies, invest-
ment banks, mutual funds,
is
employed each adult
vidual companies.
is
very
for
little
etc.,
which are
citizen can bid
The problem
is
in
turn owned by the public.
on a pseudo stock market
If
a voucher scheme
for the shares of the indi-
that there are thousands of firms to be privatized, there
information about their future profitability available, and the transaction costs
each citizen to acquire at least some information are high in comparison to the value of
his voucher. Therefore,
investment funds will enter the market and compete for the vouchers
of the citizens in return for a stake in the investment fund.
Most
of the actual bidding
the pseudo stock market will then be done by these financial intermediaries.
large
enough to
the shares of a
''For
diversify risks in their portfolios,
company
and they
will
will
be
hold a significant fraction of
so that the resulting ownership structure will not
a survey of the different privatization proposals see Borensztein and
We
They
on
Kumar
be atomistic.
(1991) and Fischer
do not consider proposals to give or sell the companies to their employees. For a discussion of
some of the problems involved with labor managed firms see Hinds (1990). Bonin (1992) discusses the actual
implementation of privatization schemes in Poland and Hungary. For the special case of Eastern Germany
see Bos (1991) and Sinn (1991). An overview of the programs in other countries can be found in RFE/RL
Research Report (1992).
(1991).
However, the organization of a voucher scheme and of the
may be very comphcated, time consuming, and
this
is
procedure
controlled
may result
by a
initial
involve high transaction costs. ^ Furthermore,
in a very concentrated ownership structure,
where a whole industry
To avoid these problems the government could
single investment fund.
distribute the firms to be privatized directly to
them "holding companies"). Their
pseudo stock market
some
portfolios could
financial intermediaries (let us call
be designed such that competitors
in
the same industry are allocated to different holdings. Furthermore, for each holding there
should be some other holding companies with similar portfolios in order to facilitate relative
performance evaluations. The holdings are owned by the general public: each adult citizen
gets an equal share of
them which he can
Both procedures allow
a
fair initial distribution of
The
market.
holding
disentangle and trade on the stock market.
for a rapid privatization of the
The
wealth.
restructuring of the
economy
is
or investment fund can
salary on the absolute
and
relative
actively monitor
and control the managers of
of his portfolio.
In particular he
up conglomerates
if
they are too
to domestic or foreign investors
A
manager
to the
manager
his portfolio.
"his" companies,
Thus he has an
of a
and to maximize the value
vertically integrated. Furthermore,
he gets an offer which
is
his
incentive to
supposed to shut down unprofitable firms, and to
much
if
left
be given an incentive contract which conditions
performance of
is
then
A
stock market should operate freely from the very beginning.
company
the company.
whole economy and start with
he
will sell
split
a firm
higher than the stock price of
of an individual firm will be actively controlled
by the
financial
intermediaries and will have additional incentives through stock options and the threat of
He
and expertise
in order to
maximize the
To ensure competitive market conditions the government should commit
to a free trade
bankruptcy.
will try to attract foreign capital
probability of survival and the profits of his firm.
policy.
As
far as non-tradables or natural
monopolies are concerned, antitrust laws and
regulatory authorities as in western economies will be necessary.
The
social costs of
unavoidable.
adjustment
will
The government should
be very high during the transition period, but they are
not interfere with the market to slow
^For a discussion of the mechanics of the voucher scheme which
(1992).
is
down the
process.
being used in Czechoslovakia see Brada
but rather deal with these costs directly by granting unemployment benefits and subsidizing
and regional development. Government revenues should come predominantly from
retraining
taxes.
.
The government approach:
Before any privatization takes place the government should undertake some coarse restructuring
itself.
Conglomerates and monopolies are broken up to avoid anticompetitive behaviour
in the future. Grossly overstaffed
tive for private investors. If a
companies are slimmed down to make them more attrac-
company
is
hopelessly unprofitable
should be shut down by
it
the government.
Again some fraction of the stock of each company may be given away or sold
at a
discount to workers and managers (this could be conditional on future privatization), and
the government could retain a minority stake for
itself.
Then the government
Some companies may be
the majority of the shares of each firm to private investors.
through
initial
public offerings on the stock market.
If
will
management buyouts
have to look actively
are also possible.
for a potential buyer,
sold
there are several investors interested in
buying the entire company the government could auction
leveraged
tries to sell
it off.
However,
in
In smaller
many
companies highly
cases the government
domestic or foreign, and to bargain with him
on the terms of privatization. To reduce the social costs of future
layoffs the
government
could insist on employment and/or investment guarantees to be given by the private owner
in
exchange
for a lower price, the cancellation of
outstanding debt, the guarantee of future
subsidies, or other privileges.^
The
is
sale of
each individual firm
will
be a
difficult
and time consuming process. There
no reliable way to estimate the market value of a company. Since these firms operated
in the past
under severely distorted
prices, trade structures
and management objectives, an
estimate of the value of the enterprise cannot be based on past experience.
sells
them too
If
the government
cheaply, it loses, potential revenues and creates "unfair" windfall profits for
the lucky investors.
If it
lack of domestic savings
bargains too hard, potential buyers
and entrepreneurs
The German Treuhand-Anstalt employs
it
will often
be
may
lose interest.
difficult to find
this policy frequently. See Sinn (1991).
Given the
a potential buyer
at
all,
in particular for the biggest
conglomerates which need a
and huge investments before they may become
Thus, the privatization process
will take
profitable.
a considerable
10 years until the majority of the large industrial companies
firms are run
directors
cis
which
is
management and
it
joint stock companies,
owned by the
state
amount
of time (at least 5 to
In the
sold).
is
meantime the
and controlled by a board of
by a government agency. The board monitors and controls the
selected
tries to
lot of painful restructuring
induce profit maximizing behavior.
If
the
company
is
unprofitable
can go bankrupt. However, the government will take the social costs of liquidation into
account and
may
subsidize the firm in order to
The revenues from
privatization
and the
smooth the adjustment
profits of the state
process.
owned companies can be
used to reduce social adjustment costs in the transition period.
The
rest
may
distributed to the population or be retained by the government to reduce current
either be
and future
taxes.
Management
3.
In this section
we take a
closer look at
what kind
Incentives
of incentives can
be given to managers under
the different proposals. This discussion prepares the ground for the assumptions
made in
the
formal analysis of the efficiency of production and restructuring under different governance
There are three
structures.
there are the managers
different kinds of
who run
the individual firms and
restructuring during the transition period.
financial intermediaries (which
arguments given below apply
for
managers that have to be distinguished:
we
Second,
if
will call "holdings"
who have
First,
to carry out the painful
the individual firms are owned by
for simplicity,
even though
all
mutual funds, pension funds, or investment banks as
the
well),
then the managers of the holding company have to monitor and control the firms in their
portfolio.
We
will call these -managers
managers on the firm
level.
Third,
if
"directors'Mn order to distinguish
the individual firms have not been privatized yet
they are owned by the state and controlled by a government agency.
agency
will
be called "administrators"
.
them from the
In the following
we
The managers
of this
discuss which kind of incentive
to each of these groups under the conditions of the transition
mechanisms can be given
period in Eastern Europe/
>
Incentives on the firm level
The standard approach
efficiently
is
to give a
manager the
right incentives to invest
and to produce
to link his salary to his performance by offering bonuses and/or stock options.
Bonuses are conditional on accounting data
(e.g.
profits, revenues,
output, etc.) and can
be used by both, privatized and state owned companies. However, they have several severe
drawbacks.
First,
is
no "baseline" of
will be.
Second, bonuses
under rapidly changing exogenous conditions there
knows what good performance
enterprise performance, and no one
do not condition on the long-term value enhancement of the
distort incentives because they
firm but rather on profits or revenues of the current period.
Thus they discourage a long-term
strategy to restructure the company. Third, in the transition period the environment of the
enterprises will be very noisy: input and output prices will change rapidly, political decisions
will
have a major impact on the profitability of many markets, and the macroeconomic
instability
imposes additional
therefore a poor measure of
modern
This noise
is
reflected in the accounting data
management performance.
Eastern European companies
deal with
risks.
is still
corporations.
in its
Finally, the accounting
which
system
is
in
infancy and there are few accountants trained to
Thus the data
will also
be inaccurate and
may be
subject
to manipulation.
Stock options can only be used
if
a
company has already been
privatized and
if its
shares
are traded on the stock market. In contrast to bonuses they do not distort incentives because
they condition on the stock market evaluation of the value of the company. However, in the
transition period they will be of limited use too. First, a well functionning stock
market does
not yet exist and will take some time to develop. Second, in the transition period the envi-
ronment
of the
companies
will
the long term prospects of the
ment. Thus, market analysts
be constantly changing which has a much stronger impact on
company than
will
variations in the performance of the
manage-
concentrate on predicting changes of the market conditions
rather than on monitoring managers. Furthermore, given the lack of reliable and undistorted
^For a more detailed survey on different incentive mechanisms and
market see the excellent discussion in Tirole (1991).
10
in particular the role of the stock
accounting data,
is
it is
also very difficult for
market analysts to assess how
efficiently
a firm
run. Thus, in the transition period share prices will contain very little information about
the performance of a manager, but rather reflect the noise of his environment. MaJcing the
manager's salary very responsive to the stock price gives only weak incentives to increase
efficiency but exposes
him
to a lot of risk.
Therefore, the
manager should be given a
flat
incentive scheme.^
Managers are
also motivated
by career concerns.
If
the managerial labor market observes
a signal about the manager's performance, this signal will be used to update the market
evaluation of the manager's ability. Thus
salary.
it
will influence his outside
Unfortunately the signal will again contain a
and there
is little
a
company can do
to
Another important device which
improve
disciplines
this
lot of
job prospects and his
noise in the transition period,
mechanism.
managers
is
the threat of bankruptcy.
If
a
firm goes bankrupt the manager loses his job and the rents and perquisites associated with
controlling the company. Furthermore, the event of bankruptcy gives a strong signal about
the manager's ability to the managerial labor market, lowering his future job prospects.
Since bankruptcy
is
a serious possibility for firms which do not adapt fast enough to rapidly
changing market conditions
to both,
in the transition period, this threat
managers of privatized and of state-owned companies.
Takeovers are only possible
argument
is
that
if
the
if
the
management
is
The standard
company has already been
privatized.
incompetent or
share prices will be lower
inefficient,
than the potential value of the firm. Thus, a raider could take
management, thereby increasing the value
in
can give powerful incentives
of the
it
over, replace the
company, and make a
profit.
western economies the role of takeovers as a device to discipline managers
controversial.^ In the transition period there
is
incumbant
However, even
is
limited and
the additional problem of a lack of domestic
®For a formal model of this argument see Holmstrom and Milgrom (1987). Milgrom and Roberts (1992, p.
221) summarize the results of the literature on performance pay in the "Incentive Intensity Principle": The
optimal intensity of incentives depends on four factors: the incremental profits created by additional effort,
the precision with which the desired activities can be assessed, the agent's risk tolerance, and the agent's
responsiveness to incentives."
Our main argument
here
is
that during the transition period the measurement
precision will be so poor that the incentive intensity has to be very small. For an interesting application of
this principle in
Ghemawat (1992).
(1988) and Schnitzer (1992) for possible negative side effects of takeovers on
a western industrialized country see
^See Shleifer and
Summers
the incentives of managers and other stakeholders.
11
savings and entrepreneurs. Furthermore, the banking system
difficult to finance
Finally,
is
hardly developed, making
it
a leveraged takeover.
managers can be monitored and controlled by the board of directors or by a
government agency. From the experience
in
western economies
it
seems unlikely that these
supervisory institutions are able to exercise a tight control of the efficiency of the manager's
operations.^" However, they can replace an obviously incompetent manager,
and they can
enforce major strategic decisions, e.g. whether or not to liquidate the firm.
Incentives for directors of holding companies
While there
is little
scope for stock market incentives on the firm
for financial intermediaries.
its portfolio,
A
holding
which reduces the
the directors of the holding to
company
volatility of its
less risk
pools firms of
own
level, this is
many
not the case
different industries in
stock price. Thus, stock options expose
than the manager of a firm. Furthermore, by having
several holding companies with similar portfolios
it
is
possible to give financial incentives
based on relative performance evaluations.^^
The
threat of bankruptcy or of a takeover, however, will play
no significant role
for
managerial incentives. These events are very unlikely given the size of the holdings.
The
directors of financial intermediaries
can be beneficial
other hand, the
on the holding.
if
may be
subject to governmental control. This
a grossly incompetent management team has to be replaced.
more the government can intervene the more
it
will
impose
its
own
On
objectives
For example, the government could press the directors not to shut
an unprofitable enterprise.
the
down
Since the losses have to be paid by the holding this policy
expropriates the shareholders of the holding.^^
i°SeeTirole(1991,p. 226).
'^Relative performance evaluations can be very useful to
filter out noise and thus to reduce the risk the
manager has to bear. See Holmstrom (1982).
'^An interesting example for such a policy is the Treuhandanstalt. Although the Treuhand is not a
joint stock company owned by East German shareholders, the profits of the Treuhand are supposed to be
distributed to East German savers, as a compensation for their losses during monetary unification. Detlef
Rohwedder, the Treuhand's first chief executive, estimated the revenues from privatization to be 600 billion
DM. However, in order to get employment and investment guarantees the Treuhand has reduced its prices so
much, that even though it has only sold the most profitable enterprises up to now, it already runs a deficit
(which will be covered by the government). See Dornbusch and Wolf (1991).
12
Incentives for administrators of a government agency
very
It is
difficult to give
efficiency.
The main
agency there
administrators of a government agency an incentive to care about
difference to directors. of a holding
no stock market, no
is
formance based compensation.
company
is
that for a government
possibility of takeovers or bankruptcy,
have career concerns and do react to
down a
per-
Administrators receive a fixed salary and will behave
bureaucrats, trying to increase their budget and their sphere of influence.
pressure not to close
and no
political pressures.
like
However, they
For example, the higher the political
firm the harder they will try to find cross-subsidies to rescue
it.
Summary
In the transition period in Eastern Europe there are only limited possibilities to give efficient
incentives to managers.
On
the firm level incentives will
come mainly from the
threat of
bankruptcy and the career concerns of the managers. Privately owned financial intermediaries
can use the stock market to base the compensation of their directors on stock options
and relative performance evaluations. In a government agency administrators
like
will
behave
bureaucrats and react to political pressure.
4.
In this section
period:
reflect
we want
to
A
Simple Model
compare two
a privately owned holding
governance structures in the transition
different
company and a government
These institutions
the two idealtypical cases of the market approach and the government approach
described in Section
2.
The
theoretical insights obtained
useful to better understand the intermediate proposals.
for the
agency.
manager
from
this analysis will also
The model
focuses on the incentives
and on the decision of the directors or
of a firm to restructure efficiently,
administrators on whether or not to keep an unprofitable firm in operation.
consider the incentive of a government agency to
be
We
do not
the companies to private investors after
sell
the transition period.
Consider a set of n firms, indexed by
i
=
13
1,
.
.
.
,n, all of
which are either controlled
by a holding company or by a government agency. There
economy which are run by
we take the
In this section
question
how
Each
different holding
n firms
n firms as given. In Section 5 we address the
portfolio of the
,
How
well a firm adjusts to
new market conditions depends on
the effort of the manager to restructure his firm efficiently.
a firm succeeds
its
its
i
We model this in a highly stylized
there are two states of the world, called "success" and "failure".
expected net present value from staying in business, denoted by
liquidation value, which
we denote by
V^^. If it fails,
value exceeds Vi}^ Thus, there are only two possible realizations of
Vj
<
Vj^
<
Vi.
by the manager.
The
A
probability of success, p,(ai),
natural interpretation
reorganize his company. His effort
if
is
the manager spends effort
quite general here.
workers, or to shut
supplier.
:
,
run by a manager. In the transition period the environment of
is
these firms changes rapidly.
way: For each firm
other firms in the
companies and/or different government agencies.
to optimally structure this portfolio.
of the
higher than
may be many
e.g.
V^, Vi
measured
G
its
{!£,,
manager
in units of utility costs for the
a,, a^
G
IR^.
VJ, is
liquidation
V,}, with
a function of an "action"
that a^ reflects the effort
he incurs the disutility
a,
Suppose
down a
is
is
is
however,
i
a,-
taken
spends to
manager,
The notion
If
i.e.
of "effort"
that the reorganization involves to lay off redundant
division of the firm in order to
These restructurings are painful
to the
buy from a more
efficient outside
manager because he has
to carry
them
through against the resistance of the workers. Or suppose that the manager has to actively
look for foreign expertise and investors. This takes time and effort, the
to a different business culture,
Assumption: The
cave in
Gi,
with
and he may
lose part of the control of his
probability of success, Pi(a,),
<
Pi{a,)
independent accross firms.
<
1
manager has
for all a,
is
to adapt
company.
increasing and strictly con-
G IR^. The
p,(a,) are stochastically
^'^
^^Note that we are not concerned about bankruptcy of the firm but rather about liquidation. A firm
which produces at average variable costs which are smaller than the market price may go bankrupt because
it cannot pay off its creditors.
However, this firm would not be liquidated. Either the banks will forgive
new-owner buys-the- firm's capital at a reduced -price, .but in both cases it is kept
these cases we define Vi as the net present value of expected future revenues minus
not including capital costs which have been commited in the past. We are grateful to Josef
part of the debt, or a
in operation.
future costs,
Brada
To avoid
for pointing this out to us.
^''Stochastic
firms does not
independence of the p,() is clearly not realistic. However, allowing
our results and does not yield any additional insights.
aflfect
14
for correlation
between
the net present value of a firm
If
maximizing owner
will shut
it
smaller than
is
down and
liquidation value, then a profit
its
sell off its assets.
In a frictionless world the assets
be redeployed to their next best use, and workers and managers
will
the going wage in other industries. In this case the owner bears
However, this
firm.
The
is
all
will find
employment
at
the costs of failure of his
not a realistic scenario for the transition period in Eastern Europe.
company may have considerable
liquidation of a large industrial
external effects which
are not taken into account by a profit-maximizing owner:
Due
-
to the high level of general
time to find new jobs
unemployment workers and managers
in other sectors,
an extended amount of time. Even
capital
Furthermore,
is lost.
if
the
if
and
their
human
resources
will
have a hard
may be left unused
they find another job, their firm specific
company dominates a town
for
human
or an entire region, part
of their physical assets (in particular real estate) will be devalued.
Outsiders to the firm
-
pany
may
also
be affected: the liquidation of a large industrial com-
affects its suppliers in other sectors of the
in liquidation,
the firm, and
it
it
may
may
economy which may
also
be forced
cause the economic decline of the town or region dominated by
lead to social unrest which increases macroeconomic and political
instability.
be useful to distinguish between workers, manager and the
It will
be the
utility loss incurred
by workers
if
company
their
i
is
rest of the society. Let
liquidated,
the respective utility loss of the manager and of the rest of society.
liquidation of firm
is
assumed
i
is
denoted by A,
to be positive
=
A|"
-f
so his
are the incentives of the
manager
wage cannot be made contingent on
is little
-|-
J"
A*.
The amount
and Af* and A* be
total social cost of
of these welfare losses
and exogenously given. Thus, from a welfare maximising point of
view, a firm that failed should be liquidated
What
A
The
Af
if
and only
of firm
a,.
i
to
if
V^ — V, >
work hard? His
A^.^^
effort is
unobservable,
Furthermore, we argued in Section 3 that there
scope for bonuses and stock options in the transition period. Thus, we assume that
the manager gets a fixed salary, the net present value of which
'^Note that the social cost of not liquidating is exactly V/' — V^.
liquidation, while its (possibly negative) profits are only Vj < V^^ if
15
is
The
it is
denoted by w.
He may have
firm could have generated V/' in
kept in operation.
an incentive to spend some
into the model, but take
minimal
level of effort, a-,
harder than
is
it
a,
=
effort
because of career concerns.
assuming that
as exogenously given,
which we normalize to
in order to
down
utility
firm
function
i
is
=
{xi
0) or
this incentive induces
reduce the probability that his firm
whether to keep
it
€
is
to
this
some
work
liquidated. If his firm
denote the decision whether to
{0, 1}
in operation (x^
=
1).
Then the manager's
given by^^
[/,
The board
of directors
particular,
we assume
is
=l£i-(l-x.)Ar-a.-
decision whether the
that the board cannot observe
company
is
shut
down
If
the firm
is
Vi,
Cj,
the effort level of the manager.
and the board takes the long-term
depend on the governance structure employed
privatized and
owned by a holding company, then
the directors of the holding dominate the board and decide whether or not to shut
an unprofitable firm.
In
or kept in operation.
incentives of the board of directors
in the transition period.
(1)
.
not involved in the day to day operations of the manager.
However, the board does observe the realization of
The
do not incorporate
The manager may be induced
0.
closed down, he incurs a utility loss Af*. Let Xj
close
We
In this case the shares of the invidual firms
down
and of the holding are
traded on the stock market. As we argued in Section 3 the directors of the holding should be
rewarded with stock options and according to relative performance evaluations.
that these incentives are powerful enough to induce
them
to shut
down a
We
assume
firm which loses
money. ^^
If
the
the board.
company
is
Since the
owned by the
company
is
state the administrators of a
government agency control
not traded on the stock market, share prices cannot be
used to give incentives. The managers of the government agency will receive fixed salaries
and are essentially bureaucrats. What are
the
money they
their objectives?
They do not want
control to the general budget, but rather want to use
it
to
to transfer
maximize
their
we have scaled a, such that the utility function of manager i is
This formulation has the advantage that we have to impose cissumptions only on the shape of
p,() but not on the shape of the effort cost function.
^'^It may be that the incentive to do so is not strong enough.
For example, the manager of the holding
^^Recall, that without loss of generality
linear in
may
a,-.
receive private benefits
from
sitting
on the board, so that he has an interest in keeping this position.
firm. Although these are serious concerns, we do
Or he may be bribed by the manager and collude with the
not consider them in the formal model.
16
sphere of influence. Thus they will try to spend the profits of the firms that succeeded to
those which failed in order to stay in control of them. Furthermore, the majiagers of the
government agency are sensitive to
political pressure.
For each firm that failed there will
be a pressure group urging the government agency to keep this particular firm
approximation we assume that the
first
So the government agency
of liquidation of each firm.^*
of liquidations.
political pressure
is
alive.
As a
proportional to the social cost
will try to
minimize the social costs
Note that the objectives of the government and of the administrators partly
coincide. If a firm fails
and the
of its future losses, then both
is
bigger than the net present value
in operation.
However, the administrators also
social cost of liquidation
want to keep
it
want to subsidize firms which should be closed down.
Can the government
interfere with the policy of its agency,
prevent wasteful subsidies?
and
losses of
If
each individual firm so that
main
we assume
can
it
it
could control cross-subsidization, then this
this information will
be incomplete.
On
grounds of
that the government has no effective control of the agency. However, our
results are unaffected
perfectly control its agency.
The time
in particular,
the government had detailed information about the profits
might be possible. In practice, however,
realism
and
if
we make the opposite assumption that the government can
We
will
structure of the model
government
decides on
privatization scheme
comment on
is
summarized
Figure
1:
in Figure
a,
success or failure
Time
structure of the
we go
along.
1.
nature
determines
manager
chooses
this possibility as
board
decides on
liquidation
model
a very rough approximation. Typically, the political pressure
is higher if there is a small but
group which is strongly affected by the decision, and lower, if the social costs of the decision
are widely spread out across society. Thus it might be more realistic to assume that the political pressure
depends only on A™ + A[", but not on A,-. This would yield an additional distortion under government
'®This
is
well organized
control.
17
,
The socially optimal allocation
.
As a benchmark consider the (unconstrained)
associated with firm
is
i
given by:
^'
Clearly,
if
no firm with
and only
if
Vi
V^ — A, >
best allocation. At date 2 social welfare
first
> V^
V^,
~ \V^-A,-ai
ifx.
should be closed down.
i.e., if
K <
If
^^^
=
K^, firm
should be liquidated
i
the expected future losses are bigger than the social costs
of liquidation.
The
a^^^
Given that
e arg
p,(a,)
is
max
{p.(a.)
•
+
F,
strictly concave,
a^^
-
(1
is
pi(a,))
•
satisfies
maxjV.^ -
A„
uniquely defined by the
p:.(aP)-min{F.-V;.^
V.}
first
-
a.}
(3)
.
order condition:
=1,
+ A.,F.-Z,.}
(4)
the manager should increase his level of effort until the marginal social benefit equals
i.e.
his
manager
welfare maximizing effort level of the
marginal
cost.
Private control: the case of a holding
The
director of the holding
him an
incentive to
firm (choose x,
=
0)
company
maximize the
if
and only
if
gets part of his salary
Thus, at date 2 he
profits of the holding.
K <
K^,
i-e.
if
from stock options which give
the firm
is
privately unprofitable.
not take into account the external effect of his decisions on society
liquidation borne by workers,
many
manager and the
will shut
(i.e.
down a
He does
the social costs of
rest of society), so there will
be
inefficiently
liquidations.
Consider now the decision of the manager at date
manager anticipates that x^
=
iff
af e arg
V^
<
max
V^^, so
{tw
-
(1
1
on how much
effort to
spend.
The
he chooses af such that
-
p,(a,)) A^*
-
a,}
(5)
aelR^^
His utility maximizing effort level
is
uniquely characterized by
p:(«f)Ar
18
=
1
(6)
,
i.e.
the manager equals marginal private benefits with marginal costs. Suppose that
A,
>
y,,
i.e.
Comparing
would be
it
(6)
and
(4)
socially efficient to shut
down the
>
and
V/'
A,-
= Af + AJ" + Af.
manager spends too
so the
little effort as
may be ambiguous. However, we
result
manager's rent
firm in case of a failure.
clear that
it is
AT < mm{V,-V/^ + Ai,V,-Vi} =
because F,
V^ —
F.-
-
Hence, concavity of
compared to the
+
^-^
A.-
(7)
that af
<
—
V^ the
p,(-) implies
best. If V^^
first
consider large companies, so
it is
most
<
A,-
af^,
likely that the
small as compared to the difference in net present values of the company,
is
so that
AT <
Thus, the manager
outcome
how
it
likely to
is
spend
-
F,
(8)
.
Although the
inefficiently little effort in this case too.^^
compared
clearly inefficient as
is
Vi
to the first best, the
more
interesting question
is
compares with what would have happened under government control.
Government control: the case of a government agency
At date 2 the government agency has to decide which firms to close down subject to the
constraint that the subsidies which have to be paid to keep negative net present value firms
do not exceed the
alive
profits
earned by profitable firms. The idea
is
that the government
agency can take the positive net present values of the profitable firms to cross-subsidize the
losses of
some companies that
be a liquidity problem
market.
We
money
This
sum
i^If
A["
to keep the
—
problem formally. One way to think about
government agency
sells
it is
to
or closes
assume
down
all
is
used to commit subsidies to unprofitable firms which are sold for an effectively
If
Vi
sum
the
— Vj
company
from
which have to be commited to keep
the
in
of the. re venues obtained
manager works too hard. Recall that we
operation after
private control the manager's rent
Vi
and there may
profitable firm can be sold for its net present value, so
of subsidies
>
losses are flows,
A
negative price.
the
this
after the transition period) the
(i.e.
and
the government agency does not have access to a perfect capital
do not consider
that at date 2
companies.
if
failed. In practice, profits
is
it
failed,
it
selling the successful firms exceeds
all
unsuccessful firms in operation,
are in the case where
but where the holding will close
lost in case of
a failure, while in the
Vj. Therefore the manager's private return to an increase in effort
19
yields a revenue Vi.
is
first
it
best
it would be efficient
down. Thus, under
CEise
the loss
is
only
bigger than the social return.
then the surplus
transfered to the state budget. ^°
is
following standard discrete linear
Thus the government agency
programming problem:
faces the
,
n
mm
mmJ2i^-Xi)Ai
(9)
»=i
^XiVi + {l-Xi)V^ >0
s.t.
.21
Ignoring the integer problem this program has a straightforward solution:
with
V,
> V^
in operation.
Index the firms that failed by
and subsidize them
—J
firms
-
is
Two
inefficient.
The budget
satisfy Vj^
V_j
socially efficient.
government agency
< Aj have been
However, since there are
outcome
-
is
budget
is
exhausted. Note that
However, the amount of subsidies paid to the
socially
more
is
exhausted before
subsidized, so too
efficient
firms which failed but
all
many
firms are closed down.
than under private control, the
less inefficient liquidations
The budet exceeds the amount
will
(10)
J+l
^j+y
cases can be distinguished:
of the
—
is
firms
€ {!,...,/}, such that
j
in the order of their indices until the
the order of subsidization
all
>,jM^.
—
^>
^]
Keep
than the one induced by the holding company.
of subsidies necessary for these firms, so
some enterprises
be kept alive although they should have been closed down. In this case the efficiency
comparison with the holding company
is
ambiguous. Note that
if
the government could
control wasteful subsidies this case would never arise.
Consider
now the
decision problem of
manager
probability p,(a,) with which his firm succeeds.
sure. In case of a failure, his firm
may
government agency has enough funds
or
may
left
i
If it
at date
His action
a,
determines the
succeeds he will keep his rent AJ" for
not be closed
to help
1.
him out
down depending on whether the
of trouble.
^°The German Treuhandanstalt is a good example for such a policy.
^^This program is a classical "knapsack" program. A computational problem may arise because x,- is
restricted to be in {0, 1}. See e.g. Dantzig (1963), ch. 26, on how this problem can be solved. The solution
is standard and not interesting in our context.
20
Given that firm
The
other firms.
i
failed there are 2"~^ possible
probability of each
outcomes of successes and
=
outcome depends on a-i
{oi,
.
.
Let Li be the set of outcomes which lead to a liquidation of firm
a-i) be the probability that firm
Prob(L,
i
will
be shut down
if it fails.
,
.
i
failures of all
a,_i ,
a,.(.i
and
let
,
.
.
.
,
«„}.
9,(a_,)
=
Then the manager's
I
problem
is
to pick a, such that
e arg max
Gi
{u;
-
[1
-p,(a,)]
•
9,(a_,)
In equilibrium the utility maximizing level of effort for
characterized by the
-
AJ"
manager
i,
a,}
(11)
denoted by af
(12) to (6)
it is
•
q,{aU)
Note that (13) holds with equality only
A- =
•
1
(12)
.
AT < AT
(13)
.
certain that the
if it is
government agency
i.
government
control.
(13) holds with strict inequality, then Pi(af )
concavity of
p,(-) implies
However,
government agency than
if
that the
In this case there
manager spends
owned by the
if it is
diminishes his incentive to restructure
5.
He
firm
is
>
Pi(af)
and
controlled by the
foresees that even
of trouble with probability (1
—
9,)
if
he
which
efficiently.
Implications of the
Both governance structures generate an
results
less effort if his
private holding.
him out
the government agency will help
is
will not
no difference between private and
have the funds to subsidize firm
The inefficiency
uniquely
clear that
qiiaU)
fails
is
,
FOC
pKaf )
Comparing
•
Model
inefficient allocation as
compared to the
"first best".
from the impossibility to perfectly align the incentives of the managers
with social welfare maximization. The manager of a privately controlled holding company
will
maximize private
of a
government agency
profits
and shut down more firms than
will shut
possible, so in this case the
down
number
the manager of an individual firm
less firms,
but he
of liquidations
is
likely to
21
is
The manager
inclined to subsidize as
is
may be
spend an
socially efficient.
much
as
too small. Under both regimes
inefficient
low amount of
effort.
But he
work harder
will
company
his
if
is
privately controlled than under the control of a
government agency. Government control leads to a
He
rationally anticipates that the
budget constraint" for the manager.
"soft
government agency
will try to help
him out
of trouble
because liquidation of his firm involves high social costs. Privatization hardens his budget
Under private control the manager knows that a private owner, who does not
constraint.
care about social welfare, will not subsidize his losses.
In the
model privatization
is
a commitment device of the government.
private control there will be no cross-subsidization from other firms.
whether
this
commitment
is
credible. If the
liquidation of a particular firm
is
the firm
happen, but
it
could step in and subsidize the losses of this firm, so that
general budget.
company by
holding
ones which
budget
is
will
and there
will
Bank not
to
going to
that these subsidies have
The government cannot expropriate shareholders
tight, the
if
these firms are
government
will
owned by the
of the
state. ^^ In the transition period the
be credit rationed on international capital markets,
be a strong pressure from international organizations
undermine macroeconomic
stability
by giving
like
the
IMF or
the World
soft credits to unprofitable firms
or by financing the budget deficit through the central bank.
is limited.'^"'
is
is
it
transferring directly the profits of successful firms to the unsuccessful
only possible
be
arises,
concerned about the social costs of
can only happen to a limited extend. The reason
come from the
under
is
The question
be closed down by the holding. There are clearly some cases in which this
will not
to
it
government
If
Thus, the scope
for subsidies
Furthermore, to the degree that subsidies from the general budget are feasible,
they would probably also be paid to the government agency. Thus, they affect both regimes
symmetrically and do not alter the basic trade-off.
How
Not
of
TO STRUCTURE THE PORTFOLIO OF FIRMS
all
them
the firms have to be privatized at the same time.
to private holding
government agencies.
If it
companies and leave the
does
so,
The government could
rest for
some time under the
which firms should be privatized
give
some
control of
first?
^^However, the government could partially expropriate stockholders by increasing the tjix rate on corporate
Goldfeld and Quandt (1991) analyze the impact of such a policy on input demands of the firm.
^•'This argument does not hold for the case of Germany. For a more subtle argument why privatization in
western industrialized countries can be a commitment device to restrict the amount of subsidies see Schmidt
profits.
(1990). For other models of the soft-budget constraint see Goldfeld
22
and Quandt (1991) and Schaffer (1989).
Our model
suggests that
private holdings.
expected future
agency there
these firms
losses, so
fail
This
it.
they should not be kept
is
manager
a
is
— V, <
and
socially wasteful
it
should go immediately to
A,-
fails
and that there are enough funds
will not hesitate to close
W\
and only
down
this firm
also gives a stronger incentive to the
it
smaller.
is
more important
for social welfare,
A,. In this case expected social welfare under private control
if
of a government
weakens the incentive of the manager to
not only socially efficient,
said about firms which are
government control
i.e.
when
higher than under
is
if
=
p,(af)F.-f(l-p,K))-[V;-^-A.]-af
>
p,(af)F.
= Wf
+ (l-p.K))-[(l-9.)Z, + 9.(K^-A.)]-af
where a^ and a^ are characterized by
1, i.e.
>
Under the control
alive.
to restructure, so the probability of failure
What can be
Vl"
This
failure.
V^-
the social costs of liquidation are smaller than the
work hard. The holding company on the other hand
in case of
—
the firms with Vl"
a positive probability that such a firm
is
to subsidize
left
If
all
(6)
and
(12) respectively.
Suppose that
qi is
(14)
close to
the probability that the government agency will have the funds to rescue the firm
small. In this case there
is little
difference
both governance structures the firm
will
between governmental and private control. Under
be closed down
manager are the same. Now suppose that
qi is
if it fails
if
the firm
fails it will
and the incentives
considerably smaller than
manager's incentives are weaker under government control, so af
However,
is
<
be cross subsidized with probability
1.
In this case the
a^ and p.(af)
(1
for the
— qi) which
<
is
Pi(af).
socially
efficient.
As can be seen from
case of success (V,)
{iii)
is
is
(14) private control
high,
if
if
V,
more
(n) the incentive effect
the net social loss of liquidation V,
straightforward. First,
is
is
large,
—
V^^
then
+
it is
likely to
is
A,- is
be optimal
strong
small.
(p,-(af)
The
if (i)
— p,(af)
the prize in
large), or
intuition for this result
very important that the firm
is
succesful.
So the manager should be given an incentive to increase the probability of success, even
this
is
hand,
only possible at the price of considerable social costs in case of failure.
if
Vi
is
small, then
it is
more important
if
On
if
the other
to prevent the social costs of liquidation rather
than to give better incentives to the manager. Secondly,
if
the action of the manager has a
strong impact on the probability of success, then his incentives are
23
more important than
in
the case where his action hardly affects the prospects of the firm. Finally,
down the
of shutting
then
it is
firm
is
more important
not
much
if
bigger than the social cost of keeping
to give better incentives to the
the social cost
it
in operation
manager than to prevent future
liquidation.
To summarize, there
-
firms with
are two kinds of firms which should be privatized immediately:
little social
significance which should not be subsidized
anyway
in case of a
failure.
-
firms for which the incentives of the
manager are very important:
effi-
by the manager has a strong impact on the probability of success
cient restructuring
or because the firm will be very profitable
On
either because
if it
succeeds.
the other hand, firms which have either very high social costs of liquidation or which
will fail
anyway no matter how hard the manager works should be kept under governmental
control.
Note that
government even
When
agencies
this will often include the firms
if
they were under the private control of a holding.
the government structures the portfolios of the holdings and of the government
revenue considerations into account.
will also take
it
pected profits
is
agencies. However,
able to cross-subsidize,
even
if
If
a
company with high
ex-
given to a holding, then these profits are lost as a source of revenue for a
government agency. This
its
which would have been subsidized by the
if
is
desirable
if
the government wants to limit wasteful subsidies of
the government
it
is
afraid that there will not be
will give this firm to
enough funds
avail-
a government agency as a source of revenue,
from the firm's point of view immediate privatization would have been
optimal.^"*
Intermediate governance structures
Two
proposals by Lipton and Sachs (1990) and Blanchard
et.al.
(1991), which have found
considerable attention in the literature, can be seen as intermediate governance structures
which yield a limited commitment of the government not to subsidize too much. Lipton and
Sachs (1990) suggest that the government agency keeps 25 to 30
^''This
may
%
of the shares which, in
be an important reason why most governments in Eastern Europe do not want to privatize
"natural" monopolies, but rather keep
them
in state
ownership and protect their profits by restricting market
entry.
24
the long run, should be sold to a core investor. Thus, only part of the profits of successful
firms goes to the government agency and can be used for cross-subsidization. This reduces
the probability that the government agency subsidizes too
constraint for the
Blanchard
et.al.
at least to
maximize
fired
owned holding companies the managers
cases
most costly
profits (through stock options),
liquidations. In
some
cases this
is
but they
Thus they
desirable.
where a harder budget constraint would have improved
of this
scheme
is
of
which
by the government. The holding managers are given some
pressure of the government which can replace them.
socially
the budget
some extend.
(1991) favor privately
and can be
are appointed
incentives to
manager
much and hardens
will also
will
respond to the
probably prevent the
However there
social welfare.
will also
be
The advantage
that the government cannot force the holding managers to cross-subsidize
too much, because of the political pressure of the shareholders of the holding
who do
not
want to be expropriated.
Conclusions
6.
The model is oversimplistic in many
respects. In particular,
possible outcomes for each firm, success
size of
the firm and the
of possible
number
how much
is
labor to employ.
and that
project for future research.
each of these outcomes the
and the manager
amount
will
continuum
have considerable
it
of workers in order to reduce the social
could subsidize an unprofitable firm to
a limited extend, so that only a fraction of the original
more
for
For example, the government agency could
and additional unemployment. Or
analysis of these possibilities with a
assumes that there are only two
given. In reality there will be a
of directors
force a profitable firm to keep an inefficient
costs of layoffs
failure,
of employees
outcomes and the board
discretion about
and
it
company
will survive.
detailed and complicated
model
is
A
thorough
an important
However, we believe that the basic trade-off derived in our
very simple model will carry over to more complex cases: government control will lead to
more
cross-subsidization and less social costs than private control, but
effects
it
will
have adverse
on management incentives. Managers anticipate that the budget constraint under
government control
is soft,
so they will spend less effort to restructure efficiently.
25
References
Blanchard,
"Reform
Bos, D.
O., R.
in
Eastern Europe"
Brada,
L.
Summers
(1991):
Cambridge, Massachusetts.
Press,
Germany:
A
Survey of Current Issues", mimeo.
The Worker's Stake
"Privatization and Efficient Contracts:
in the Tran-
Journal of Comparative Economics, Vol. 16(4).
Borenzstein, E. and M.
rope",
MIT
Krugman, R. Layard, and
Monetary Fund.
J. (1992):
sition",
,
P.
"Privatization in East
(1991):
International
BONIN,
Dornbusch,
IMF Staff Papers,
J. (1992):
Kumar
"Proposals for Privatization in Eastern Eu-
(1991):
Vol 38., No.2 (June), 300-326.
"The Mechanics
of the
Voucher Plan
in Czechoslovakia",
RFL/RL
Re-
search Report, Vol. 1(17), April 24, 42-45.
Dabrowski,
J.,
M. Federowicz, and a. Leviatas
(1991):
"Polish State Enterprises
and the Properties of Performance: Stabilization, Marketization, Privatization",
and
Politics
Society, Vol. 19, 403-437.
Dantzig, G.
(1963):
Princeton,
New
''Linear
Programming and Extensions" Princeton University
,
Press,
Jersey.
Dornbusch, R. and H. Wolf
(1992): "East
German Economic
Restructuring", mimeo,
MIT.
Fischer, S. (1991): "Privatization
Paper No. 3703.
Ghemawat,
Easter European Transformation",
in
NBER
Working
P. (1993): "A Case Study in Organizational Efficiency: Competitive Position
and Internal Organization", Journal of Comparative Economics, forthcoming.
GOLDFELD,
S.
AND R. QUANDT
(1991): "Effects of Bailouts, Taxes,
and Risk-Aversion on
the Enterprises", Journal of Comparative Economics, Vol. 16, 150-167.
Hinds, M.
cialist
(1990):
"Issues in the Introduction of
Market Forces
in
Eastern European So-
Economies", Internal Discussion Paper, The World Bank, Washington.
HOLMSTROM,
B. (1982): ""Moral Hazard
in
Teams"", Bell Journal of Economics, Vol.
13,
324-340.
HOLMSTROM,
B.
AND
P.
MiLGROM
"Aggregation and Linearity in the Provision
(1987):
of Intertemporal Incentives" ^canojnetrica. Vol. 55, 303-328.
,
KORNAI,
J. (1980):
''Economics of Shortage"
KORNAI,
J. (1990):
"The Road
LiPTON, D. AND
J.
Sachs
to a Free
,
North Holland, Amsterdam.
Economy", Norton, New York.
(1991): "Privatization in Eastern Europe:
Brookings Papers of Economic Activity, 293-333.
26
The Case
of Poland",
,
MiLGROM,
AND
P.
Englewood
J.
Cliffs,
ROBERTS
New
RFE/RL Research Report
por^,
Vol.
(1992):
^Economics, Organization and Management"
Jersey: Prentice Hall.
""Radio Free
(1992):
Europe/Radio Liberty Research Re-
1.
Sappington, D. and
J.
Schaffer, M.
"The Credible- Commitment Problem
Stiglitz (1987): "Privatization, Information and Incentives",
Journal of Policy Analysis and Management, Vol. 6, 567-582.
(1989):
lationship", Journal of Comparative Economics, Vol.
Schmidt, K.
SFB
(1990):
"The Costs and Benefits
in the Center-Enterprise Re-
13, 359-382.
of Privatization", Discussion
Paper A-287,
303, University of Bonn.
Schnitzer, M.
(1991): "Breach of Trust in Takeovers
ter", Discussion
and the Optimal Chorporate Char-
Paper, MIT.
Shleifer, a. and L.H. Summers (1988): "Breach of Trust in Hostile Takeovers",
Corporate Takeovers: Causes and Consequences, ed. by A.J. Auerbach, London.
Shleifer, A. and R. Vishny (1992):
in:
First Steps",
mimeo,
East Germany", Working Paper No. 3998,
NBER,
"Privatization in Russia:
Harvard.
Sinn, H.-W. (1991): "Privatization
in
Cambridge.
TiROLE,
J.
(1991):
Transition",
"Privatization in Eastern Europe:
NBER Macroeconomic Annual,
\
5939 073
27
221-259.
Incentives and the Economics of
Date Due
Lib-26-67
MIT LIBRARIES DUPl
3
TDflD
DDflEMmM
M