Is Employee Ownership Unstable

Is Employee Ownership Unstable? Some Lessons From a Transition Economy[i]

Panu Kalmi

Helsinki School of Economics, department of economics

Abstract: Estonia is one of the many post-communist economies, where the ownership of many state enterprises was transferred to their employees after the collapse of communism. On the basis of a large sample of Estonian enterprises, I found out that the number of enterprises owned and controlled by the employees was rapidly declining as the economic transition progressed. This supports the argument that employee-owned firms cannot survive in competitive markets, and that they will inevitably "degenerate" into conventional firms. However, we must carefully take into account the institutional factors that led to this outcome. The Estonian evidence does not indicate that employee-owned firms are bound to fail everywhere, but it does tell under which conditions degeneration is likely to occur. I conjecture that similar developments may be underway in many other transition economies.

1 Introduction

The question whether employee-owned organisations are able to survive in market economies is almost as old as the co-operative movement. Especially supporters of co-operatives were disappointed about the failure of worker co-operatives to survive and flourish. The possibility that worker-owned and controlled co-operatives would degenerate into conventional, capitalist enterprises was discussed already by John Stuart Mill in his Principles of Political Economy. These ideas were developed further by the Russian economist Mikhail Tugan-Baranovsky and the British socialists Sidney and Beatrice Webb.[ii]

Since the 1970s, there has been conducted empirical studies related to the issue of degeneration. Degeneration has been defined as conversions to capitalist firms, increase in the number of "hired" (i.e. non-owning) employees, or outsider dominance of control structures. The studies have found mixed results. Employee-owned firms may survive for a long time and the rate of business failures among them is not any worse than that of conventional firms. Conversions occur to some extent, but not even close to the extent predicted by the most dismal theorists. The use of hired labour is relatively common, but there exists quite much variation among the different types of firms. In general, the degeneration theories have proven to be more relevant for the case of US Pacific plywood co-operatives (where the membership fee is market-determined) and for Israeli industrial (Histadrut) co-operatives, whereas, for instance, the French SCOP co-operatives and Mondragon co-operatives in Spain have proven to be quite resistant against degeneration. Owner (member) control has prevailed especially in co-operatives, while the record is worse for US ESOPs in this regard. The reason for the relatively low density of enterprises that are employee-owned and -controlled seem to be more related to the low entry rates of these enterprises rather than to high exit rates.[iii]

On the basis of this evidence, one might expect that degeneration merely a theoretical problem that does not plague employee-owned enterprises in reality. However, the Estonian evidence indicates that this is not so. Employee ownership is rapidly declining in Estonia. In order to understand the reasons for this, it is necessary to review the privatisation process that led to the spread of employee ownership in Estonia. This is done in section 2. In section 3 I discuss how the decline of employee ownership takes place in practice, who takes over the employee-owned firms, and what are the reasons behind the decline. In chapter 4 I discuss the general lessons of the Estonian case on the stability and viability of employee ownership.

2 Employee ownership in Estonia

The roots of employee ownership in Estonia are in the last reform attempts of Soviet Union, known as perestroika. When the inefficiencies of economic planning and state control of all sectors of the economy became ever more apparent by the mid-1980s, the first secretary of the communist party of the USSR, Mikhail Gorbachev, decided to allow the formation of privately-owned enterprises. Between 1987-1990, several thousands of small state enterprises and worker co-operatives were established in Estonia. In relation to the size of the economy (Estonia had only 2 mln inhabitants), the density of these small private enterprises was in Estonia one of the highest among the socialist republics of USSR. The workforce in small state enterprises and co-operatives gained the right to decide on production, pricing, wage setting, and investment, although in the limits imposed by state authorities. Importantly, they obtained also the right to keep the revenues. However, it became soon apparent that these measures were insufficient to revive the economy. In 1989 started semi-privatisation of large industrial enterprises, known as leasing of state enterprises. The leaseholders gained the right to control the enterprise and the right to residual revenues. In essence, they became private firms, lacking only the third important element of ownership, namely the right to transfer the assets to third parties. They also had to pay rent to the government for the use of assets.

The most important sources of employee ownership in Estonia were however the programmes launched by the Estonian authorities: the Estonian variant of leasing that started in 1990, and the privatisation of collective farms that started in 1992. (Estonia became fully independent in 1991.) Since the centralised privatisation programme of large enterprises started only in 1993, leasing was for the time being one of the few options for the state firms to separate itself from the state control. The lease contract necessitated that a significant proportion of employees became co-owners. Unlike in the Polish and Russian variants of leasing, the leaseholders did not have the pre-emptive right to buy the assets when the lease contract expired. However, it was extremely rare that someone else than the leaseholder (i.e., employees collectively) bought the property. Of all state enterprises privatised before 1994, around 40% were transferred to the majority ownership of non-managerial employees and an additional 25% to managerial ownership.[iv] However, in autumn 1993 a new privatisation programme was adopted, where it was stipulated that the enterprises should be sold in a tender to strategic investors. The Estonian Privatisation Agency selected the winner of privatisation contest on the basis of the price offered for the shares and on the development prospects the new owner could offer for the enterprise. After this change in policy, it became very rare that the enterprises were privatised to employees.

The privatisation of collective farms to their employees was based on the principle that the property belonged to the employee collective and not to the state. Similar principle was also used in the privatisation of state farms, although the state was a nominal owner of these farms. The employees were granted "labour shares" according to the length of their service in the collective farm, which they could invest in the enterprise assets. The collective and state farms consisted of several units, most of which were auxiliary to agricultural production (e.g. small manufacturing, electricity works, construction etc.) Although there were only around 340 collective and state farms in 1989, it has been estimated that over 3000 successor enterprises existed in 1995, of which a substantial number were employee-owned.

According to our estimates, in around 25 % of all privatised enterprises non-managerial employees were the dominant owners in 1995. This already reflects a decline from the time of privatisation. However, in 1999, the number of firms that were employee-owned was only around 12%. In 80% of firms with some employee ownership in 1995 the percentage of equity belonging to employees had diminished, and in another 80% of them the participation ratio (number of owners in the workforce to the total number employees) had declined. Finally, while two-thirds of employee-owned firms changed ownership in the four-year period from 1995 to 1999, employees gained control in only around 5% of the other enterprises; that is, not only are employees losing control in initially employee-owned enterprises, but employee buy-outs remain rare.[v]

3 Decline in employee ownership

There are in principle three ways how employee ownership can diminish: Through employees selling their shares, through new share issues, or through attrition, when employees leave the firm but keep their shares. Many observers of transition economies have believed that employee ownership would be only temporary, since employees would rush to sell their shares at the first instance when it is possible. However, the survey data I collected on Estonian firms having at least some degree of employee ownership, indicates that, at least for Estonia, the view that employees sell their shares quickly is not correct.[vi] Shares are often traded in small quantities. In average, around 4% of the shares are traded annually. However, this average hides quite substantial variation, a few observations with substantial trading increasing the mean. In around 50% of the surveyed firms there were no shares traded during the last three years. When there is some trading of shares, it is usually the retired employees who are selling their shares, not those employees who are still employed by the firm. To be sure, sometimes employees sell all their shares to managers or outsiders, but such cases occur more seldom than is commonly thought. Furthermore, trading of existing shares is almost certainly not the most common way of changing ownership structures. Especially managers are in favour of increasing their shareholding by issuing new shares. Employees become suspicious if managers want to buy their shares and might demand a large premium for their shares. However, they are generally less interested in participating in new share issues, where managers can buy shares at low nominal prices and increase their shareholding and control over the enterprise.

However, the most common reason for the decline of employee ownership and ownership change is attrition. Virtually none of the studied enterprises have any mechanism for including new employees as owners. When old employees retire and keep their shares, and new employees do not become owners, the proportion of equity owned by employees as well as participation ratio falls, and the number of outsider ownership grows. Almost all companies surveyed employee ownership declined because of attrition. In (at least) 15% of firms that were employee-owned in 1995, the largest group of owners in 1999 were former (retired) employees. Sometimes attrition has decreased the level of employee ownership so much that managers or outsider owners have become dominant owners. In around 25% of employee-owned companies managers have become the dominant owners, and the number of employee-owned firms taken over by outsider investors is of similar magnitude.

Detailed case studies give further evidence on the extent of this problem. I surveyed 160 employees in five employee-owned enterprises. One of the aims of the survey was to found out the determinants of ownership at the level of the individual. It appeared that one variable could predict exceptionally well the status of ownership: whether the respondent was employed by the firm before privatisation or not. Of the respondents who were employed before privatisation, three in four were owners in their firms. In contrast, of those respondents who had joined their firm after privatisation, only one in seven was also an owner. While some other variables could also explain ownership (higher education, higher position in the enterprise hierarchy, and longer tenure also beyond the qualititative effect described above), it is clear that employment status at the time of privatisation was the most decisive factor. This points out that Estonian employee-owned enterprises fail to extend ownership to new employees.

What are the reasons behind this failure? In my view, there are four main reasons. Perhaps the most important is at the micro (enterprise) level: There are simply no sufficient incentives to maintain or extend the level of employee ownership. Employee ownership was viewed as a device to separate enterprises from the state control, to earn profits and to achieve greater job security than under investor ownership. However, employee ownership was not perceived as an instrument to improve productivity or to democratise labour relations. Consequently, employee-owned firms do not show higher productivity than other firms, neither does employee ownership improve job attitudes.[vii] Therefore, the incumbent employees saw little reason to diminish their own share in profits or endanger their own job security by taking in new employee-owners. It was profitable to them to take new employees in rather as "hired labour", without ownership status.

Second reason was the lack of support at the government level. It is indeed quite surprising that employee ownership became so wide-spread, even though no serious political force in Estonia supported it. Privatisations to employees took place in the absence of centralised privatisation programme; it happened "by default" rather than by design, and it was tolerated rather than supported. However, this also meant that the government did nothing to support the viability and stability of employee-owned firms.

Third, employee ownership took the form of direct, investment-based individual share ownership. It is well-known that direct share ownership is an unstable form of ownership, especially when compared with employment-based approaches, such as co-operatives or ESOPs.[viii] The lack of market for shares has somewhat slowed down the pace of degeneration, but this has not been sufficient.

Fourth, the level of employee representation in Estonian enterprises is incredibly low, in employee-owned enterprises as well as in other enterprises. The Soviet-era trade unions lost their legitimacy in the systemic change, and the successor unions are very weak. The right-wing governments have not been supportive to trade unions. The Estonian case once again shows that employee ownership requires real employee representation in order to succeed.

There is also a fifth possibility: that employee ownership fails because it is inefficient. However, the studies on the performance of employee-owned enterprises in Estonia show that while employee-owned enterprises fail to outperform other types of enterprises, neither are they performing significantly worse, so this can hardly explain a significant amount of ownership change.[ix]

4 Conclusion: can employee ownership survive?

Do these findings mean that employee ownership is a stable ownership form, because if failed in Estonia? No, but the Estonian case can teach us a great deal what employee-owned firms need in order to succeed. First, employee ownership has to arise on voluntary basis. No enterprise in Estonia was forced to become employee-owned by the government. However, most of them became employee-owned at the time when the choice was between continued state ownership and employee ownership. Had the choice been broader, many managers would no doubt had chosen otherwise. This explains why many managers diluted the ownership of employees by issuing new shares in the first possible instance, and why employees themselves did not often bother to defend their rights as owners. Second, there should be strong employee organisations. There are bound to become conflicts between managers and non-managerial employees, especially when managers themselves are significant owners. Because of the informational advantage of the managers and their hierarchical position, employees cannot successfully solve these conflicts without representative organisations, such as unions. These organisations are, moreover, helpful in solving the conflict of interest that might sometimes arise among employees. Third, there is now good evidence that employee ownership is most likely to succeed when there are supportive organisations (e.g. networks of employee-owned firms) and when the state action is supportive, or at minimum not harmful.[x]

However, I think that the most important (but perhaps most controversial) lesson to be drawn from the Estonian experience is that direct share ownership is not consistent with sustainable employee ownership. This is mainly because under direct share ownership, it is difficult to establish a mechanism that guarantees that shares are transferred from retiring employees to new employees. A number of reasons related to asymmetric information and risk aversion prevent the efficient functioning of market for shares in employee-owned enterprises. Co-operatives or ESOPs would do much better in this regard. However, co-operatives gained an unearned bad reputation under communism. Many people associate them with the old regime and especially with the hated collectivisation process. Moreover, a transformation based on wide-spread application of co-operatives would certainly have been deemed too radical, not only by Estonian right-wing governments, but also by international development organisations such as the World Bank or EBRD. A better solution could have been the use of ESOPs with well-developed mechanisms of including new employees.

However, one should note that in Estonia, as well as in many other transition economies, employee-owned enterprises were viewed as inefficient and only as a temporary stage towards full "capitalist" ownership structures, dominated by large strategic investors and foreigners. According to this view, the decline of employee ownership is not "degeneration", but progress. However, my view is that this argument is misplaced. It is not clear that ownership changes contribute to greater technical efficiency, and in some cases clearly they do not. For instance, it can hardly be argued that an ownership structure where former employees form the majority of owners is an efficient one.

It can be reasonably argued that the possibility to change ownership structures is important in order to allow efficiency-enhancing ownership transfers. However, it is possible to construct ownership structures that allow simultaneously for sustained employee ownership and substantial outsider involvement in ownership. In my view, a promising way of doing this is by having two classes of equity. The first class of equity is owned by the employees collectively. The revenue rights are distributed among employees in some fashion decided by employees (e.g. by reflecting differences in tenure or salary, or, alternatively, equally among employees). The second type of equity would be tradable. Employees could, if they wish, to sell to outsider investors even a number of shares enabling to majority control of the enterprise.[xi] In this way, decision-making rights would shift outside the enterprise only when employees collectively decide to transfer these rights. This would be a great improvement over the present situation, where such decisions are done in effect by the management.

Another question is how relevant all this is at the present state of economic transition. The transition cannot be reversed, and the decline of employee ownership in Estonia will continue, perhaps even at an accelerated pace. My rather pessimistic forecast is that in 10-15 years there will be very little employee ownership left in Estonia.

Unfortunately, there is only limited evidence on these issues from other transition economies. The scant available evidence there is from Russia and Poland suggests that similar developments are under way in these countries as well.[xii] This is not surprising given that in these countries, employee ownership was based on direct ownership of shares, as in Estonia. It is possible that in some other transition economies where the chosen property form was closer to ESOPs (Romania and Hungary) employee ownership might have been of more permanent nature. However, from the Estonian experience we can draw the general lesson that direct share ownership is not sustainable.

References:

Blair, M., D. Kruse and J. Blasi (2000): 'Is Employee Ownership an Unstable Form? Or a Stabilizing Force?' In M. Blair and T. Kochan (eds.): The New Relationship: Human Capital in American Corporations, Washington D.C., Brookings.

Blasi, J.R., M. Kroumova and D. Kruse (1997): Kremlin Capitalism: Privatising the Russian Economy, Ithaca: Cornell University Press.

Bonin, J.P., D.C. Jones and L. Putterman (1993): 'Theoretical and Empirical Studies of Producer Cooperatives: Will the Twain Ever Meet?', Journal of Economic Literature, 31:1290-320.

Ellerman, D.P. (1990): The Democratic Worker-Owned Firm, Boston: Unwin Hyman.

Jones, D.C. and N. Mygind (1999): 'The Nature and Determinants of Ownership Changes After Privatisation: Evidence From Estonia', Journal of Comparative Economics, 27: 422-41.

Jones, D.C. and N. Mygind (2001): 'Ownership and Productive Efficiency: Evidence From Estonia', Review of Development Economics.

Kalmi, P. (2002): On the (In)stability of Employee Ownership: Evidence From Estonia and Lessons for Transition Economies, Ph.D. dissertation (manuscript), Center for East European Studies, Copenhagen Business School.

Kozarzewski, P. and R. Woodward (2001): ‘Secondary Privatisation in Poland (Part I): Evolution of Ownership Structure and Company Performance in Firms Privatized by Employee Buyouts’, CASE Reports No. 47, CASE Foundation, Warsaw.

Russell, R. (1985): Sharing Ownership in the Workplace, Albany, NY: SUNY Press.

Russell, R. (1995): Utopia in Zion: The Israeli Experience With Worker Cooperatives, Albany, NY: SUNY Press.

Smith, S.C. (2001): 'Blooming Together or Wilting Alone? Network Externalities and the Mondragon and La Lega Cooperative Networks', Discussion Paper Nr. 2001/27, WIDER, United Nations University, Helsinki.

Toscano, D.J. (1983): 'Towards a Typology of Employee Ownership', Human Relations, 7:581-602.