June 1, 1999    
This Date's Issues: 3315 •3316 • 3317  3318 

Johnson's Russia List
#3317
1 June 1999
davidjohnson@erols.com

*******

World Bank
Annual Bank Conference on Development Economics
April 28-30, 1999
Keynote Address
Whither Reform? Ten Years of the Transition
By Joseph E. Stiglitz
Senior Vice President & Chief Economist 
World Bank

Paper prepared for the Annual Bank Conference on Development Economics,
Washington, D.C., April 28-30, 1999. The findings, interpretations, and
conclusions expressed in this paper are entirely those of the author and
should not be attributed in any manner to the World Bank, to its affiliated
organizations, or to the members of its Board of Directors or the countries
they represent.

Table of Contents
Introduction
Part I: Misunderstandings of Market Economies
Competition and Privatization
Privatization Alternatives
Creative Destruction
Social and Organizational Capital
The Post-Socialist Separation of Ownership and Control
Reducing Agency Chains: Stakeholder Privatization
Restructuring and Bankruptcy
Restructuring through Decentralization, Reconstitution, and Recombination
Part II: Misunderstandings of the Reform Process
Sequencing and Pacing of Reforms
The Grabbing Hand of the State; the Velvet Glove of Privatization
The Modern Debate: Shock Therapy versus Incrementalism
The CDF and The Presumption for Participation
Conclusion
References

Abstract

Ten years after the beginning of the transition, what are the lessons to be
learned? Broadly speaking, most observers would conclude that China's path
to the transition has been a success so far, while Russia's path has not
been. I argue that the failures of the reforms in Russia and most of the
former Soviet Union are not just due to sound policies being poorly
implemented. I argue that the failures go deeper, to a misunderstanding of
the foundations of a market economy as well as a misunderstanding of the
basics of an institutional reform process. For instance, reform models
based on conventional neoclassical economics are likely to under-estimate
the importance of informational problems, including those arising from the
problems of corporate governance; of social and organizational capital; and
of the institutional and legal infrastructure required to make an effective
market economy. They are also likely to underestimate the importance of
the creation of new enterprises-and the difficulties of doing so. The
promise, for instance, of quick economic transformation, and the creation
of a "people's capitalism," based on voucher privatization with investment
funds has proven illusory. An alternative strategy of decentralization,
pushing economic decision-making down to the level where the stakeholders
can protect their own interests without presupposing elaborate legal
machinery that will take much longer to evolve, may under the circumstances
prove to be more effective. Given the choice between the momentum of
bottom-up popular involvement in "flawed" reforms and top-down imposition
of what reformers see as "clean model institutions," an argument can be
made in favor of using our knowledge and experience to work to improve the
bottom-up approach to transformation. 

The varied experiences of the countries going through the process of
transition represents one of the most important set of economic and social
experiments ever conducted, and should provide a rich opportunity for
researchers both to understand the process of reform and to gain insights
into the workings of economies. The limited success in so many of the
countries means that their remain many opportunities for applying the
lessons of such studies. 

Whither Reform?: Ten Years of the Transition
Joseph E. Stiglitz

Introduction

This century has been marked by two great economic experiments. The
outcome of the first set, the socialist experiment that began, in its more
extreme form, in the Soviet Union in 1917, is now clear. The second
experiment is the movement back from a socialist economy to a market
economy. Ten years after the beginning of the transition in Eastern Europe
and the Former Soviet Union: How do we assess what has happened ? What
are the lessons to be learned? Surely, this is one of the most important
experiments in economics ever to have occurred, a massive and relatively
sudden change in the rules of the game. As rapidly as the countries
announced the abandonment of communism, so too did western advisers march
in with their sure-fire recipes for a quick transition to a market economy. 

A decade after the beginning of the transition in Eastern Europe and the
Former Soviet Union (FSU), and two decades after the beginning of the
transition in China, the picture is mixed. Each country started the course
of transition with a different history, a different set of human and
physical endowments. Some had lived under the yoke of central planning and
authoritarianism for most of the century, while in others it was imposed
only in the aftermath of World War II. Those countries bordering Western
Europe with encouraging prospects of European Union integration were
clearly in a different position than the land-locked countries of Mongolia
and the former Soviet republics in Central Asia. Counterfactual
history-what would have been but for the policies that were pursued-is
always problematic, and no more so than when there are so many variables
with which to contend. Yet, the disparity between the successes and
failures is so large that it calls out for interpretation and explanation,
and in any case, the public debate has already begun. 

The contrast between the strategies-and results-of the two largest
countries, Russia and China, may be instructive. Figure 1 shows that over
the decade beginning in 1989, while China's GDP nearly doubled, Russia's
GDP almost halved; so that while at the beginning of the period, Russia's
GDP was more than twice that of China's, at the end, it was a third
smaller. Not only did Russia stagnate during this past decade, but Figure
2 shows how it succeeded in turning the theoretical tradeoffs of inequality
and growth on its head - in the process of shrinking its GDP, Russia also
doubled its inequality (as measured by the Gini coefficient). Recent data
contained in the 1999 World Development Indicators paint an even bleaker
picture, with poverty-defined as $4 a day-rising from 2 million to over 60
million by the middle of the decade.

The titles of some recent books by leading advisors in the transition
process are telling-How Russia became a Market Economy or The Coming Boom
in Russia. While those who had advised Russia on its transition path
constantly predicted that it was on the verge of success-virtually
declaring victory just a short while before its recent crash-the shortfall
should have been apparent. Yes, Russia had succeeded in "privatizing" much
of its industry and natural resources, but the level of gross fixed
investment-a far more important sign of a burgeoning market economy-has
fallen dramatically over the last five years [EBRD 1998]. Russia was fast
becoming an extractive economy, rather than a modern industrial economy. 

Standing in marked contrast with these failures has been the enormous
success of China, which created its own path of transition (rather than
just using a 'blueprint' or 'recipe' from western advisors). It succeeded
not only in growing rapidly, but in creating a vibrant, non-State-owned
collective enterprise sector. While investment in manufacturing in Russia
stagnated, that in China was growing by leaps and bounds. Critics of such
comparisons point out the marked difference in starting points-China's
income was far lower, and so there were more opportunities for catch-up.
On the contrary however, I would argue that China's challenges were
greater, for it had to manage the challenges of transition and of
development simultaneously. China did better than countries of comparably
low income, while the countries of the FSU and Eastern Europe, by and
large, did worse than countries of comparable income.

The question that we need to ask is, why the failures? Not surprisingly,
those who advocated shock therapy and rapid privatization argue that the
problem was not too much shock and too little therapy, but that there was
too little shock. The reforms were not pursued aggressively enough. The
medicine was right; it was only that the patient failed to follow the
doctor's orders! Other defenders of the recommended reform programs argue
that the failures were not in the design of the reforms, but in their
implementation. One of the Russian reformers recently quipped that there
was nothing wrong with the laws they enacted except non-enforcement. 

I want to argue here however, that the failures of the reforms that were
widely advocated go far deeper-to a misunderstanding of the very
foundations of a market economy, as well as a failure to grasp the
fundamentals of reform processes. I will argue below that at least part of
the problem was an excessive reliance on textbook models of economics.
Textbook economics may be fine for teaching students, but not for advising
governments trying to establish from anew a market economy-especially since
the typical American style textbook relies so heavily on a particular
intellectual tradition, the neoclassical model, leaving out other
traditions (such as those put forward by Schumpeter and Hayek) which might
have provided more insights into the situations facing the economies in
transition. (But, as I shall argument below, the failings of textbook
economics are far greater: with few exceptions, they fail even to
incorporate insights concerning corporate governance, a concern of
mainstream economics from Marshall [1897] to Berle and Means [1932] and a
major focus of modern information and transactions costs economics)
A part of the problem also rose from confusing means with ends: taking, for
instance, privatization or the opening of capital accounts as a mark of
success rather than means to the more fundamental ends. Even the creation
of a market economy should be viewed as a means to broader ends. It is not
just the creation of market economy that matters, but the improvement of
living standards and the establishment of the foundations of sustainable,
equitable, and democratic development. 

Finally, while due obeisance was paid to "political processes"-and insights
into the political process were often put forward in justification of
particular courses for reform-in fact, little understanding of these
political processes was evidenced. In hindsight, it is clear that many of
the political forecasts of those involved in the reform process were far
from clairvoyant; many worries seem, by and large, not to have
materialized, while political developments which should have been of
concern were not anticipated. Nor can one separate "principles" from how
they are, or are likely to be implemented. Policy advisers put forth
policy prescriptions in the context of a particular society-a society with
a particular history, with a certain level of social capital, with a
particular set of political institutions, and with political processes
affected by (if not determined by) the existence of particular political
forces. Interventions do not occur in a vacuum. How those recommendations
are used, or abused, is not an issue from which economists can simply walk
away. And this is especially so in those instances where one of the
arguments for the economic reforms is either failures in the political
process or their impact on the political process itself. It is time for
the doctors to rethink the prescription. But in doing so, they will have
to take the patient as he is today, not as he might have been had history
taken a different course. The point is not to refight the old battles, but
to learn the lessons of the past, to help guide the future. 

Part I: Misunderstandings of Market Economies

In my book, Whither Socialism? [1994], I argued that the failure of market
socialism arose in part from a failure to understand what makes a actual
market economy function-a failure arising in part from the neoclassical
model itself. If the Arrow-Debreu (AD) model [1954] had been correct, then
market socialism might have fared far better. But while the AD models
capture one essential aspect of a market economy-the information conveyed
by price signals, and the role that those price signals serve in
coordinating production-the information problems addressed by the economy
are far richer. Prices do not convey all the relevant information. I want
to suggest here that those advocating shock therapy, with its focus on
privatization, similarly failed because they failed to understand modern
capitalism; they too were overly influenced by the excessively simplistic
textbook models of the market economy. But we should be less forgiving of
those failures. While Hayek and Schumpeter had earlier in the century
developed alternative paradigms, views that had not been well integrated
into the mainstream of the Anglo-American tradition, by the time the
post-socialist economies faced their transition, the modern theory of
information economics had shown the striking limitations of the AD model,
and used the tools of modern economic analysis to illustrate forcefully the
problems of corporate governance that Marshall [1897], Keynes [1963
(1926)], Berle and Means [1932], Galbraith [1952], March and Simon [1958],
Baumol [1959], Marris [1964], and many others had written about over the
course of the twentieth century. 

In the following paragraphs, I want to review what I see as the major ways
in which, for want of a better term, I shall refer to as the "Washington
Consensus" doctrines of transition, failed in their understandings of the
core elements of a market economy. In the second section of this talk, I
shall focus more sharply on the reform strategy, that is, on views about
sequencing and pacing.

Competition and Privatization 

Standard neoclassical theory argues that for a market economy to work well
(to be Pareto efficient), there must be both competition and private
property (the "Siamese twins" of efficient wealth creation). Both are
required, and clearly, if one could wave a magic wand and instantaneously
institute both, one would presumably do that. The issue however, concerns
choices: if one cannot have both, should one proceed with privatization
alone?

While those pushing for privatization pointed with pride to the large
fraction of state enterprises that were turned over to private hands, these
were dubious achievements. After all, it is easy to simply give away state
assets, especially to one's friends and cronies; and the incentives for
doing so are especially strong if the politicians conducting the
privatization exercise can get a kickback, either directly or indirectly as
campaign contributions. Indeed, if privatization is conducted in ways that
are widely viewed as illegitimate and in an environment which lacks the
necessary institutional infrastructure, the longer-run prospects of a
market economy may actually be undermined. Worse still, the private
property interests that are created contribute to the weakening of the
state and the undermining of the social order, through corruption and
regulatory capture.

Consider the incentives facing the so-called oligarchs in Russia. They
might well have reasoned: democratic elections will eventually conclude
that their wealth was ill-begotten, and there will thus be attempts to
recapture it. They might have been induced to pursue a two-fold strategy:
on the one hand, to use their financial power to gain sufficient political
influence to reduce the likelihood of such an event; but, assuming that
that strategy is inherently risky, to use the other hand to take at least a
significant part of their wealth out of the country to a safe haven.
Indeed, the "reform" advisors facilitated this process by encouraging-in
some cases even insisting-on the opening of capital accounts. Thus, the
failure of privatization to provide the basis of a market economy was not
an accident, but a predictable consequence of the manner in which
privatization occurred.

Privatization Alternatives

Those advocating rapid privatization faced the quandary that there were no
legitimate sources of private wealth within the country with which
privatization could be accomplished. Governments thus faced essentially
four alternatives-a sale of national assets abroad; voucher privatization;
taming "spontaneous" privatization; or what I shall call for want of a
better term, "illegitimate" privatization. The latter was the route Russia
chose after 1995 in the notorious "loans-for-shares" scheme. The
government can allow private entrepreneurs to create banks, which can lend
these private parties money with which to buy the enterprises (or in the
loans-for-shares deal, lend to the government, with the shares of
government enterprises as collateral). Whoever got the banking license got
a license to print money, and the license to print money is a license to
acquire government enterprises. While the corruption was somewhat
roundabout-and the process was less transparent than if the government had
simply given the nation's assets to its friends-there is in fact little
distinction between the two processes. 

Since the whole process was widely viewed as illegitimate, this "robber
baron" privatization put market capitalism to even greater disrepute than
perhaps the indoctrination of the Communist era. And since there was no
presumption that those who thereby acquired the assets were particularly
good managers, there was no reason to hope that the assets would be better
deployed than they had previously been. To be sure, some who advocated
this process worried little about either the political impact or the
managerial incompetence: they believed that there were strong incentives
in play for an "aftermarket"-so that the assets would eventually be sold to
those who could best manage these enterprises. The hope was that these new
"robber barons" would at least conduct a good auction. But this process
failed for several reasons: first, there remained the underlying
problem-where were the internal managerial teams with the requisite
capital? Worse still, the declining confidence in the economy and the
government made the country even less attractive to foreign investors. The
oligarchs found that they could extract more wealth from asset stripping
than from redeploying assets in way that would provide the foundations of
wealth creation.

The voucher schemes proved little more successful, with the Czech Republic
(at first taken as a model) providing the clearest illustration of the
underlying problem of corporate governance, the public good of corporate
management, which I will consider later.

Perhaps trying to discipline spontaneous privatization might have offered
the greatest hope: breaking up the large enterprises into smaller units
which might have provided a basis for more effective governance by the
stakeholders, a possibility I turn to later in this essay.

Creative Destruction 

An essential part of the transition to a more efficient economy is the
redeployment of resources from less productive to more productive uses.
Moving workers from low productive employment to unemployment does not, by
itself, increase productivity. Indeed, productivity is lowered, and some
productivity is better than none. The movement into unemployment is a
costly and inefficient intermediate stage-one could only defend it if there
were no better way of moving workers directly from a low productivity job
to a higher productivity job. A crude form of Say's law (with little
empirical basis) was often put forward: a large supply of idle workers
will create a demand, partly by facilitating downward pressure on wages.

But any student of the process of enterprise creation and entrepreneurship
would have expressed concerns, especially in regions like FSU, where there
was little history of market-oriented entrepreneurship. For
entrepreneurship to succeed, certain skills need to be developed in
practice, skills which those in the FSU had no opportunity to develop.
They had acquired skills in evading government regulations, in arbitraging
away some of the inefficiencies in government regulations for private
profit, and in operating at the interstices between the legal and illegal
world. But that is a far different kind of enterprise that creating new
businesses and competing in the international market place.

Entrepreneurship also requires capital. As noted above, few had the
necessary capital-especially after inflation eroded what little savings
people had accumulated. The banking system had no experience in screening
and monitoring loans, it was wrong even to think of these banks as "banks"
in the western sense. The language here confused both those in the country
and western advisers. But in any case, few of the enterprises actually got
into the business of providing funds to new, small enterprises, and thus,
even under the best of conditions, entrepreneurship was stifled. Where
then, were the new jobs for those forced out of existing employment?

Bankruptcy, or the credible threat of it, is a crucial part of a market
economy. The institution of bankruptcy, like its inverse of
entrepreneurship, had little or no precedent in the socialist countries.
The institution of bankruptcy had to be created. A variety of available
models for bankruptcy codes had evolved over centuries in the market
economies, and each was integrated into the specifics of its economy. A
transplant to an alien environment could hardly be expected to quickly take
root - particularly in the absence of an independent and competent
judiciary, trained in and sympathetic to the basic tenets of bankruptcy.
Those who hoped that newly drafted and "installed" bankruptcy codes would
drive industrial restructuring have been much disappointed. 

Moreover, as is so often the case, there is no "one best way"; all
bankruptcy systems involve genuine tradeoffs between creditor and debtor
rights. Systems need to be tailored to the local environment. For
instance, one relevant feature is the speed with which assets can be
re-engaged in productive use. In countries with little entrepreneurship,
poor social safety nets, and little tradition of labor mobility, we must
expect a tilt towards debtor-oriented bankruptcy. Moreover, we should not
think that much industrial restructuring will come out of bankruptcy
courts; the real restructuring is usually done to keep companies out of
formal bankruptcy. I will later consider a range of such restructuring
actions.

Entrepreneurship and bankruptcy, entry and exit, must be seen as two sides
of the same coin of economic change. The advice to "just enforce the
bankruptcy laws" or "just harden the budget constraint" is not good advice
where there is little culture of new business creation. Both parts of
Schumpeter's phrase "creative destruction" must be remembered. Even
long-standing market economies do not get out of deep depressions, where
many firms qualify as bankrupt, by forcing large numbers of firms into
bankruptcy. Vigorous programs of employment creation and maintenance,
through promotion of entrepreneurship and/or by Keynesian stimuli, must go
hand in hand, if not precede, bankruptcy-induced restructuring.

Social and Organizational Capital

It has long been recognized that a market system cannot operate solely on
the basis of narrow self-interest. The informational problems in market
interactions offer many chances for opportunistic behavior. Without some
minimal amount of social trust and civil norms, social interaction would be
reduced to a minimum of tentative and distrustful commodity trades. Behind
these social norms stands the machinery of the law which itself stands
apart from the market.

Property systems are in general not completely self-enforcing. They depend
for their definition upon a constellation of legal procedures, both civil
and criminal. The course of the law itself cannot be regarded as subject
to the price system. The judges and the police may indeed be paid, but the
system itself would disappear if on each occasion they were to sell their
services and decisions. Thus the definition of property rights based on
the price system depends precisely on the lack of universality of private
property and the price system. ... The price system is not, and perhaps in
some basic sense cannot be, universal. To the extent that it is
incomplete, it must be supplemented by an implicit or explicit social
contract. [Arrow 1972, 357]

The information requirements for, and transactions costs involved in,
implicit and explicit contract enforcement are typically different, so that
the two should best be thought of as complements rather than substitutes.
The problem in the economies in transition was that both enforcement
mechanisms were weak: the state's legal and judicial capacities were
limited, while the very process of transition-high institutional turnover,
high shadow interest rates, and short time horizons-impairs the
effectiveness of implicit contracts. Thus, even if institutions did not
need to be created, the very process of transition provides impediments to
the workings of a market economy.

Arrow, Hirschman [1992], Putnam [1993], Fukuyama [1995], and others have
argued that the success of a market economy cannot be understood in terms
of narrow economic incentives: norms, social institutions, social
capital, and trust play critical roles. It is this implicit social
contract, necessary to a market society, that cannot be simply legislated,
decreed, or installed by a reform government. Some such "social glue" is
necessary in any society. One of the most difficult parts of a
transformation, such as the transition from socialism to a market economy,
is the transformation of the old "implicit social contract" to a new one.
If "reformers" simply destroy the old norms and constraints in order to
"clean the slate" without allowing for the time-consuming processes of
reconstructing new norms, then the new legislated institutions may well not
take hold. Then the reforms will be discredited and the "reformers" will
blame the victims for not correctly implementing their ill-considered designs.

One variation on this theme is to blame the failure of the shock therapy
reforms on corruption and rent-seeking at every turn (e.g., Åslund 1999
this conference. But while rent seeking and corruption were important,
there was more to the failure than that (and indeed, if rent-seeking were
the sole problem, then the reductions in rents asserted by Åslund should
have been accompanied by a soaring of national output.) Moreover,
corruption and rent seeking may itself have been increased by the manner in
which the reforms were conducted, which both destroyed the already weak
social capital and which enhanced opportunities and incentives for such
activities.

The social and organizational capital needed for the transition cannot be
legislated, decreed, or in some other way imposed from above. People need
to take an active and constructive role in their self-transformation; to a
large extent, they need to be in the driver's seat. Otherwise the reform
regime is only using bribes and threats to induce outward changes in
behavior insofar as behavior can be monitored-but that is not transformation.

In market economies, firms may be seen as local non-market solutions to
collective action problems where transaction costs inhibit coordination by
market contracts [Coase 1937]. In the new post-socialist market economies,
as in the established market economies, the primary example of extensive
(i.e., beyond the family) social cooperation in daily life is found in the
workplace. Thus entrepreneurial efforts that arise out of or spin off from
existing enterprises may be particularly effective in post-socialist
societies in preserving "lumps" of social and organizational capital. Once
dissipated, organizational capital is hard to reassemble, particularly in
environments with little entrepreneurial experience. Other social
organizations that might incubate and support entrepreneurial efforts
include local township governments, unions, schools, colleges, cooperatives
(housing, consumer, credit, and producer co-ops), mutual aid associations,
guilds, professional associations, churches, veterans' associations, clubs,
and extended family groups. Creativity and experimentation should be the
order of the day to remobilize social resources particularly in the
slow-starting transitional economies of the former Soviet Union.

The Post-Socialist Separation of Ownership and Control

Given the difficulties in reassembling organizational capital once it is
dissipated or destroyed, it is particularly important to promote
entrepreneurial restructuring in existing enterprises. The need for
fundamental enterprise restructuring has not been lost on the swarms of
western advisors, but their advice has sometimes contributed as much to the
problem as to the solution. 

In retrospect, one of the remarkable features of the body of western advice
given to post-socialist economies ("Washington Consensus"), especially as
they approached the issue of privatization (see above), is the absence of
attention to the separation of ownership and control. The intellectual
framework often seems to be a curious pre-Berle-Means world where "private
ownership" and control of the enterprise are essentially the same thing-as
if the small or medium-sized closely-held corporation was the norm. Yet
the salient feature of the large companies in the Anglo-American economies
has been what Berle and Means called the "separation of ownership and
control." Keynes, even earlier, made the same point.

One of the most interesting and unnoticed developments of recent decades
[written in 1926] has been the tendency of big enterprise to socialize
itself. A point arrives in the growth of a big institution-particularly a
big railroad or big public utility enterprise, but also a big bank or a big
insurance company-at which the owners of the capital, i.e., the
shareholders, are almost entirely dissociated from the management, with the
result that the direct personal interest of the latter in the making of the
great profit becomes quite secondary. [Keynes 1963, 314]

The divergence of interests between the managers and shareholders in large
publicly-traded corporations has been a major source of the economics of
agency contracts. Yet the hard lessons of the separation of ownership and
control, and the resulting agency problems, have received insufficient
attention in the standard western advice in spite of much discussion of
"the corporate governance problem." Let me give you a few examples of
"fine phrases."

"Clearly Defined Private Property Rights"

Instead of trying to control managers in state-owned companies with better
incentive contracts, the standard advice is to privatize and let "private
property rights" provide the natural incentives-"like in the West." Yet
the separation of ownership and control in large western companies means
that the control function is not allocated on the basis of "clearly defined
private property rights." The ownership of shares, like the ownership of
bonds, is indeed clearly defined; the shareholder can buy, sell, or hold
those rights. But those rights do not "add up" to a real ownership-based
control of the company when the shareholders are atomized and dispersed.
One way to express this point is to recognize that the management of a
publicly held company with disperse ownership is a public good-and as in
the case of any public good, there will be an undersupply. Another way of
putting the same point is that the market for managers-the process of
take-overs-is highly imperfect, and does not in general ensure that the
company will be managed by those who will ensure that the assets yield the
highest returns.

"Controlling Private Owner"

When the problems of dispersed shareholding were recognized in operating
companies, then the suggested solution was usually to have a "controlling
private owner" in the form of an investment fund-as in the standard form of
voucher privatization promoted by the Washington Consensus and modeled
essentially on the Czech voucher privatization program. One obvious
problem with this "solution" is that the voucher investment funds had an
even greater "corporate governance" problem than the companies in their
portfolio. The funds' shares were held by a broad cross-section of the
entire population of citizens. Thus the shareholders' influence on the
fund management was essentially nil. Yet the controlling investment fund
idea was "sold" by the standard Washington Consensus as a "solution"
(rather than a worsening) of the corporate governance problem.

"Natural Incentives of Private Ownership"

In economics, as in politics, it is a good idea to "follow the money." Who
has the economic interests ("cashflow rights") normally associated with
corporate ownership? The standard theory is that the economic interests
are attached to share ownership. The shareholder enjoys the economic
return to ownership in two ways: dividends on shares and capital gains on
shares when sold. But when there is a separation of ownership and control,
then the controlling agent is partly or almost wholly disconnected from
those "natural incentives of ownership." The effects of the separation are
aggravated when there is pyramiding. 

An actual case is point is the Czech voucher privatization scheme. The
voucher investment funds were limited to at most a 20% stake in a portfolio
company, and the funds were controlled by fund management companies
receiving 2% of the asset value under management. That fund management
company's economic interest in the portfolio company is 0.4 % (20% x 2%).
If two funds with the same management company each have a 20% interest,
then we have a 0.8% economic interest. Other variations allowed 30%
maximum holdings and 3% fund management fees for a 0.9% economic interest.
Moreover, these returns are gross to the fund manager. If the fund manager
must expend any costs, say in devising and implementing a restructuring
plan for a portfolio company, then those costs would have to be subtracted
to get the net return to the fund management company.

Let me ask you. If you had control of an economic asset but could only
extract say 0.9% through a certain channel, would you tend to use that
channel or to find a more "efficient" channel for extracting value? At
least in retrospect, no one should be surprised that the Czech investment
funds found other channels to extract or "tunnel" value out of the
portfolio companies. After all, one does not need a lot of sophisticated
economics to figure out that if the controlling interest must pay over 99%
tax on money taken out one door, then there will be a determined search for
another door to get the money out. 

"Better Management Contracts"

It will be said, perhaps the answer lies with better regulations and
well-designed incentive contracts for the controlling fund managers. But
let us now step back and take stock. If a government had such incredible
monitoring and enforcement powers to overcome such disincentives, why not
apply those powers directly in corporatized state-owned enterprises and
then privatize later in a better thought-out way? 

One of the main points of "privatization" was to use the "natural
incentives of private ownership" instead of the more contrived incentives
of management contracts (e.g., in SOEs). Yet we have come full circle. We
have seen that the rapid privatization schemes promoted by the standard
western advice (voucher privatization with investment funds) did not
establish or lead to controlling owners motivated to restructure
enterprises towards long-term economic success. The current advice has
ended up focusing on better regulations and management contracts to try to
get those in control (e.g., the Czech fund management companies) to act
like "private owners"-since the standard form of privatization didn't do
that job. It is time to rethink that "standard" type of privatization.

Reducing Agency Chains: Stakeholder Privatization

A modern market economy is based on highly developed agency relationships.
One of the most important ways in which real economies diverge from
textbook models is in the problems of asymmetric information, imperfect
monitoring, and opportunistic behavior. Accordingly some of the most
important economic institutions arise to alleviate agency problems (e.g.,
the legal machinery to enforce shareholder rights and other stakeholder
rights, liquid stock markets and open-end investment funds (so investors
can "vote with their feet"), the legal framework of competition policy, the
entire monitoring system of accounting and auditing, and lastly the ethos
of managerial professionalism). In the more stable and developed market
economies, long multi-stage chains of agency relationships have developed
(e.g., workers are agents for managers who are agents for shareholders such
as mutual funds whose shares are held by pension funds which act as agents
for their beneficiaries such as workers). But in earlier stages of
development, market economies had much shorter agency chains. 

These agency institutions need to incrementally grow and evolve over
decades. If one tries to just set up a market economy overnight with such
extended and concatenated agency relationships, then the superstructure may
collapse in dysfunction. That is what has happened in Russia and the
former Soviet Union. The elites who have had the roles of institutional
agents representing broad constituencies in the FSU have, in many cases,
not been able resist grabbing what they can. The elites have betrayed
society's trust in them on a massive scale. Those who would enforce the
agency relationships and other legal obligations are too often themselves
part of the problem.

That is why it is time to rethink the elaborate agency chains we have been
trying to "install" in the former Soviet Union. Think for a moment why we
condemn oligarchs and managers for asset stripping and looting that leads
to the demise of an enterprise. One might say that they are within their
legal rights as shareholders. Yet we nonetheless condemn them because of
the direct impact on the livelihoods of workers and indirect impact on the
economic life of the local community and on the prospects for related
parties such as suppliers and customers. By bringing in the interests of
these other parties in evaluating the looting, we are in effect identifying
these other parties as stakeholders in the enterprise. The stakeholders
all have implicit contracts signifying long term relationships with the
enterprise. It is these stakeholders who are ultimately harmed as an
"externality" by the agents' betrayal of the extended agency relationships
in the transitional economies.

If the pyramided agency relationships are not functioning and it will take
many years to build the supporting institutions, then perhaps it is best to
shorten the relationships so that those who are monitoring are the ultimate
stakeholders who are hurt by the looting and malfeasance. Instead of A
trying to get B to get C to do something for A, the pyramided agency
relationships should be shortened as much as possible. Shortening one
stage means A trying to get B to do something for A. The most dramatic and
self-enforcing arrangement is the unification of principal and agent so
that A helps him- or herself directly. Then "corporate governance" becomes
a more manageable problem, if not a solved problem. There is no corporate
governance problem with unified principal and agent in the family farm or
small owner-operated business. In general, we might reason that the
shorter the agency chain, the easier it is to resolve the corporate
governance problem.

This is a strategy of privatization to stakeholders which could be seen as
a way to generalize the owner-operated business or family farm to
medium-sized and larger firms [see below on decentralizing large firms].
Since the stakeholders, by definition, have a long-term economic
relationship to the enterprise, they have broader interests in the firm and
another channel through which to exert their corporate governance on the
management. Their cooperation is necessary for the firm to function so
this "hold-up power" gives the stakeholders a way to exercise "corporate
governance" as part of their day-to-day business relationships rather than
through external legal machinery. They are not unrelated absentee
shareholders who see the enterprise only as a "property" (perhaps to be
quickly harvested) or who are dependent on agency chains and intermediary
institutions to exert their influence.

This general strategy would push towards decentralization. The idea is to
push decision-making responsibility down to the levels where people can
more directly control their agents or where peer-monitoring can
operate-without presupposing the elaborate institutions of monitoring and
enforcement that will take many years to develop. There is usually
corruption also at decentralized levels but centralization keeps control
too removed from the discontents that can lead to change. As David
Lilienthal, past Chairman of the Tennessee Valley Authority, put it:

[C]entralization to avoid unsavory local influences surely deprives the
people of the chance to draw their issues locally and to clean up their own
local inadequacies. The fundamental solution is to crowd more, not less
responsibility into the community. Only as the consequences of
administrational errors become more localized, can we expect citizens to
know which rabbit to shoot. [Lilienthal 1949, 89-90]

This strategy would also entail energizing some of the more subdued
segments of the population such as the workers and their unions. If
oligarchs and/or managers are looting an enterprise and destroying people's
future jobs, then any national pride in being "long suffering and enduring"
is quite misplaced. Those who are being hurt should have the information
and the organizational capacity to vigorously protect their interests, not
just to depend on some reform elite to act in their best interest. The
same holds for local governments as well as suppliers and customers-all of
whom are stakeholders in the enterprise being looted. The cooperation of
the stakeholders is necessary for the enterprise to function and the
interests of the stakeholders are harmed when the enterprise is looted.
Therefore stakeholder privatization coupled with stakeholder empowerment
will tend to reunite the de facto control rights and de jure ownership
rights in a self-enforcing system of corporate governance. Since the
strategy of shortening agency chains leads in the direction of devolution
to ultimate stakeholders, let us step back and look at the role of
bankruptcy and decentralization in restructuring-particularly in the larger
firms.

Restructuring and Bankruptcy

Industrial restructuring to improve competitiveness has proven one of the
most difficult and intractable parts of the transition process. Hopes that
privatization would lead to restructuring "by the market" have been widely
disappointed. Part of the blame, as I have noted, should be assigned to
privatization methods that created little incentive for restructuring as
opposed to "tunneling" value out of firms. But part of the blame should
also be assigned to a failure to understand the nature of the restructuring
process in the context of economies in transition. 

One fundamental error (similar to one which we have encountered in the past
couple of years in East Asia) is a failure to distinguish between what is
required in the case of restructuring a single firm within a well
functioning market economy, and restructuring virtually an entire economy,
or at least the manufacturing sector of an economy.

Systemic reorganization is different from reorganization when a single firm
has a problem. When a single firm is restructured in an economy operating
at full employment, firing underemployed workers has beneficial effects,
partly because those workers get quickly redeployed to more productive
uses. When, however, there already is massive unemployment, firing workers
moves them for a situation of underemployment to no employment-not
necessarily a transfer that leads to an overall increase in the
productivity of the economy, though it may improve the balance sheet of a
particular firm. In the economies in transition, many bad investment
decisions had been made. The issue facing a country, given its capital
stock, was not, in the short run, whether it wished it had a different
capital stock; there was no magic wand by which it could convert its
inefficient steel mills into efficient aluminum smelters. The question
was, given its capital stock, what was the best way to employ its workers.
To be sure, even in the short run some rearrangements in the labor force
were desirable; some firms should be hiring workers, some firing workers:
this is an on-going part of the process of any dynamic economy. But it
cannot be the case that all firms are overstaffed-except if at the same
time new firms can be and are being created to absorb the workers let go.

It is particularly important to recognize that there was no reason to
believe that the inherited financial structure of the firms in the economy
in transition at the beginning of the process had any inherent "merit,"
simply because finance under the socialist regime played a completely
different role. Banks did not have the job of screening and monitoring. A
high debt-equity ratio-leading a firm to a situation where it could not
meet its obligations-thus had no informational value; it did not convey
information even about the incompetence of the chief financial officer.

By the same token, when there is systemic bankruptcy, selling off assets
may make little sense: who is there to buy them? And even when it is
possible that some firm could more effectively utilize the asset, if
capital market imperfections are rampant-so that the firm that should be
expanding does not have access to capital-it may not find it possible to
obtain the funds to purchase the asset. The rearrangement of assets in the
presence of a systemic problem is thus far more difficult than in the case
of an isolated weak firm. 

Thus, financial restructuring was necessary, but a weak financial position
had far different implications than in an ongoing market economy; and the
prospects of fundamental improvements in the underlying structure of the
economy from disposing of assets, in the presence of systemic bankruptcy,
were far bleaker. 

Restructuring through Decentralization, Reconstitution, and Recombination

There was a form of restructuring that was important-but unfortunately,
with few exceptions, such restructuring was not the focus of attention-and
that is restructuring to decentralize decision-making. Indeed, it seems to
me that there is a rather general model of restructuring which suitably
describes successes in a wide variety of contexts. Consider a centralized
organization that is encountering consistent failure in its tasks. It
could be a unit of government or an economic enterprise. The "iron law of
oligarchy" has done its work so the organization is centralized, ossified,
and stagnant. Those who have power in the organization want to maintain
its structure to preserve their power and perquisites. 

New and complex situations call for experiments; not one but many
experiments. The need for many parallel experiments to see "what works"
implies decentralization so that the smaller units can operate with some
independence. Risk is diversified since a bad decision by one unit does
not entail the same for the other units. Decentralization means vertically
and/or horizontally dis-integrating a large firm into separate
semi-autonomous teams or profit centers within a federal structure-or
perhaps a split into more independent business units (e.g., spin-offs which
could be confederated and/or partly owned by the mother firm).
Decentralization can improve incentives (by linking actions of individuals
and smaller units to rewards) and accountability, and harden budget
constraints-eliminating the cross subsidies that often exist in large
organizations. 

New managers are needed in the decentralized parts. This devolution is the
hardest part of the process since it entails the central management giving
up a good part of its power to the decentralized or spun off units under
younger middle managers. Yet it is key. Restructuring for a market economy
entails a sea-change away from the strategy of keeping or conglomerating
together the largest units possible to be more successful in lobbying for
subsidies. Instead of the slogan "United we stand; divided we fall," the
slogan might be "Centralized, we go to the ministry; decentralized, we go
to the market."

One can argue that it is best if the pressure for the center to cede power
to the decentralized units to begin the process of re-constitution come
from the constituent stakeholders (e.g., workers, creditors, and other
parties with stable relationships to the enterprise), those who will lose
if the organization is not successfully restructured. Their participation
and involvement in the restructuring decisions will lead to better
execution of the restructuring plans (since the stakeholders will have more
"buy-in" and "ownership" of a plan they helped devise).

The fundamental fact is that much of the relevant knowledge is, in fact,
decentralized; that the well recognized failures of central planning-having
at least in part to do with the inability of the central planner to gather
and disseminate relevant information-can apply with equal force within a
large organization; and that simply calling the organization a "firm" does
not by itself provide the constituent parts with incentives to transmit
information to the center, nor does it endow the "firm" with the ability to
process that information, nor does it provide a clear mechanism for the
central headquarters to convey instructions to its constituent parts, nor
does it provide the incentives for those parts (and the individuals within
them) to respond in the desired manner.

Following the decentralization, the new units can experiment to probe their
environment, to test their capabilities, and to accumulate local knowledge.
The connection or loop between experiment and feedback is, under
decentralization, now much tighter so the learning process can proceed
apace. Benchmarking and world quality standards provide a Hirschmanian
pacing mechanism to drive the learning. Real decentralization within an
enterprise means that the units can now buy supplies and sell outputs
outside the firm whereas before they were in effect restricted to a
monopoly supplier or buyer within the firm. It also means the new units
should bear the costs of their failure just as they may reap the fruits of
their success. These competitive possibilities will expose vulnerabilities
within the various units to induce learning and change. Thus
decentralization along with the benchmarking and outside competition (like
"export promotion" in the international arena) could be seen as social
learning mechanisms to drive the process of recombination and
restructuring. Thus, the horizontal discussions between the units should
be seen not simply as "best practice fora" but as part of the
"constitutional" process of rebuilding the organizational relationships
from the ground up. This is the process of rebuilding social capital, to
which so little attention was paid in the process of transition.

We have so far illustrated the general model of restructuring through
decentralization, reconstitution, and recombination by applying it to a
large and presumably distressed firm. The model helps to explain why
successful restructuring is so rare in post-socialist countries (as well as
elsewhere). The center refuses to decentralize power to start the new
process of learning and reconstitution. The center clings to the hope that
some new master restructuring plan (coupled no doubt with "new machinery"
as a technological fix) will solve the problem. If the government is to
foster restructuring in troubled firms, then it needs to find ways to
promote restructuring through decentralization, reconstitution. Where
enough constituent units have the leadership (e.g., middle managers) and
initiative to strike out on the restructuring path, then the government can
help to devise ways to lift the dead hand of the center (i.e., the failed
managers clinging to centralized power) so the process of renewal can go
forward.

[continued in Part II]

********