What would you do if you had to privatize over 7,000 medium- and large-scale enterprises quickly, but you had no idea what they might be worth? Moreover, what if you were in a formerly communist country where household savings rates were low, less than 1 percent of production was generated by the private sector, and there were no significant domestic capitalists who could buy such enterprises? Finally, what if simply auctioning assets would result in foreigners acquiring firms cheaply, thereby thwarting your efforts to build a “culture of capitalism” among your citizens who had lived for decades under central planning? This is the policy dilemma that the Czechs and Slovaks faced in the immediate aftermath of the 1989 “Velvet Revolution,” which inaugurated the transition to democratic rule. (The analysis throughout refers to the federation that existed between the Czech and Slovak Republics until end-1992, unless otherwise stated.)
The solution they came up with was an extremely innovative and controversial one. In effect, they created a new method of privatization: mass privatization. It overcame the absence of domestic savings by giving away shares to the public, and it overcame the problem of how to value enterprises’ assets—one of the biggest stumbling blocks to privatization in transition economies—through a system of sequential auctions. Although it is too early to judge whether the scheme enhanced the productivity of enterprises, it is clear that the scheme met its basic objectives, without undue costs, and demonstrated the usefulness of sequential auctions for generating information about enterprises’ true market values.
How to privatize?
The former Czechoslovakia started the transition to a market economy as one of the most extreme cases of a centrally planned economy, with about 98 percent of property in state hands as recently as 1989. There were about 7,000 medium- and large-scale enterprises and about 25,000-30,000 small-scale enterprises in the state sector. Almost by necessity, the government adopted a multi-track approach that relied on a variety of privatization methods, including: (1) restitution to original owners; (2) small-scale privatization through public auctions; (3) transfer of state property to municipalities; (4) transformation of cooperatives; and (5) privatization of medium- and large-scale enterprises through direct sales, public tenders, joint ventures, and the mass privatization scheme.
Government officials generally stuck with conventional methods of privatization in cases where these methods would result in a rapid transfer of ownership. But for the vast majority of medium- and large-scale firms another approach was needed, as conventional methods would have resulted in long delays and generated fears that public assets would be sold cheaply to foreigners or to a few rich nationals to the detriment of other citizens. Thus, policymakers opted for a mass privatization program for almost three quarters of the medium- and large-scale enterprises undergoing privatization at the start of the reform program.
Nemat Shafikfrom Egypt, is a Senior Economist in the Bank’s Chief Economist’s Office, Middle East and North Africa Region, and was the Country Economist for the Czech and Slovak Republics. She has degrees from the London School of Economics and Oxford.
The enterprises in the “first wave” consisted of 1,491 firms, for a total book value of $10.6 billion (a “second wave” is going on in each Republic, albeit with different approaches after the break-up of the federation). The sectoral distribution mirrored the entire economy, with the largest number in the engineering and building and civil engineering sectors, followed by food processing, design, agriculture, and domestic trade. The average labor force was 850 workers, with a minimum of 6 workers and a maximum of 34,231. As other privatization methods, such as direct sales, grow in importance, the proportion of total assets privatized through the coupon scheme is likely to diminish to about one third of total book value.
The program commenced with the government publishing the list of enterprises to be privatized as well as detailed information about each firm—including book value, financial performance, output, employment levels, other shareholders (if any), and debts—to enable citizens to make informed choices. The government did not attempt to restructure enterprises prior to privatization, but instead focused on rapid preparation of enterprises for sale to private owners who would be best placed to restructure the firms. This approach was consistent with international experience that efforts at government-led restructuring are rarely worthwhile—the government’s expenditures on restructuring tend not to be recouped when the firm is eventually sold.
All citizens over the age of 18 were entitled to “coupons” that allowed them to bid for shares in enterprises to be privatized. Over 80 percent of eligible citizens (equivalent to 8.5 million people) registered for coupons, paying a nominal fee of about $35 that covered all the administrative costs of running the scheme. In exchange, citizens received a coupon of about $1,250 in book value terms (reflecting the total book value of enterprises to be sold in the first wave divided by the number of people who purchased coupons). Of those citizens who participated, almost three quarters gave their coupons to one of the 429 private investment funds that had emerged independently to participate in the privatization program. These funds, which operate like mutual funds, played a crucial role in “market-making” because they had done considerable homework on the value of firms (see box). Since citizens had the option of turning over their coupons to the investment funds, they were able to self-select into two groups: informed and uninformed bidders. The uninformed bidders tended to give their coupons to an investment fund to manage, thereby boosting the proportion of information relative to “noise” in the market.
There were five bidding rounds. Initially, all shares were offered for the same price, but prices quickly diverged as the market responded to changes in demand—by the third round the spread between the highest and lowest price shares was 776-fold. By the final rounds, there was evidence that equilibrium prices had emerged—93 percent of shares on offer had been sold and 99 percent of the coupons distributed had been used.
How the auctions fared
An assessment depends crucially on whether one believes that “ownership matters.” Judged by the government’s own objective—transforming ownership through a fast and fair process—the scheme was successful. The fairly narrow objective reflected the view that ownership transformation alone was sufficient to achieve productivity gains in enterprises. While there is growing evidence that this transformation does result in improved performance, virtually all the evidence comes from market economies where privatization was part of an incremental, rather than a fundamental, effort. Thus, any economic assessment for a transition economy must also consider the consequences for corporate governance—whether the diffuse ownership that emerges reduces the likelihood that shareholders will take an active role in restructuring and overseeing the enterprise.
Speed. On this ground, the scheme was successful. The entire cycle of preparation, public information, and bidding for 1,491 enterprises took 14 months—an average of over three medium- and large-scale enterprises privatized per day. The economic gains from such speed are that there is less scope for assets to deteriorate when public enterprises remain in “ownership limbo” and a greater likelihood of rapid restructuring.
Equity. Equity was achieved by giving all citizens equal opportunities to obtain shares through a transparent process. There was no systematic bias in share allocations in favor of “insiders” (either management or workers), in contrast to subsequent mass privatization schemes in other transition economies (such as Russia and Poland). Even so, some equity issues did arise. Firms were required to proceed with privatization, regardless of declared insolvency, to avoid management using financial distress as an excuse for avoiding privatization. In some cases, these insolvent firms were very large or politically important and sometimes sold for high prices, reflecting expectations of a government bailout. However, any such bailout would have resulted in either windfalls to existing shareholders or to “renationalization” of the enterprises.
Corporate governance. The average firm emerged with about half its shares owned by investment funds (averaging three major funds per firm) and half owned by individual citizens. Thus, the investment funds can be expected to play a key role in enterprise restructuring, with the secondary market serving as an important way for them to consolidate their holdings (although they are restricted to owning no more than 20 percent of the shares in any enterprise). Will this structure of ownership help? On the plus side, anecdotal evidence of funds taking an active role in the supervision and control of enterprise managers and in the replacement of non-performing managers in some enterprises is encouraging. But the funds will also be under pressure to distribute dividends, which may result in decisions inconsistent with long-term restructuring. Eventually, there is likely to be a differentiation across the spectrum between funds wanting to operate like passive mutual funds, with small stakes in a diversified portfolio, and funds that will operate like venture capitalists, taking large stakes and activist roles in management.
Learning through bidding
Before the first round of bidding in the mass privatization program, there were two types of information available to bidders: (1) public information in the newspaper about enterprises’ characteristics and past performance, and (2) private information that individuals and funds had about enterprises’ prospects. Econometric analysis of the determinants of prices in the bidding rounds reveals that the former clearly mattered, especially in the early rounds when private information had not yet been revealed. Firms that sold for higher prices tended to be characterized by high profits, strategic investors (i.e., investors taking significant shareholding stakes), and small and highly productive labor forces. In one famous case, a typographical error resulted in an extra zero being added to the profits of a hotel being offered. At the end of the first round, demand for shares in the hotel were about 400 times the supply of shares—implying that people did rely on published information to guide their bids.
However, this published information could never explain more than 29 percent of the variation in firms’ ultimate equilibrium share price. Rather, information revealed through the bidding process (such as a share’s price in the previous round or the deviation of a share’s price from the average in a particular round) held the key to eventual equilibrium prices. This “market information,” reflecting forward-looking assessments, could explain 85 percent of the variation in prices by the final rounds.
Share prices increasingly behaved like a “random walk”—implying that past prices were the best predictor of future prices—which is consistent with expectations about how an efficient asset market would function. There is also evidence that citizens’ bids tended to be correlated with bids by investment funds in earlier rounds, implying that citizens recognized and used the superior information possessed by the funds.
Arguably, under all mass privatization schemes, effective “secondary markets” for share trading will improve the initial valuation of enterprises’ shares and send important signals about corporate governance. But the experience in several transition economies indicates that secondary markets sometimes take a long time to become active, often because information about firms is not available and few market participants want to trade in ignorance. Thus, mass privatization programs that use information effectively are likely to have a head start on establishing more efficient asset markets, encouraging better corporate governance, and creating more productive private firms.
The mass privatization approach has now been copied and adapted in other transition economies—such schemes have taken place, are underway, or are under preparation in Bulgaria, Kazakhstan, Lithuania, Mongolia, Poland, Romania, and Russia. Each program reflects the objectives and constraints faced in each country. In a large country like Russia it was not possible to have simultaneous, nationwide bidding rounds, so auctions were held throughout the year at the local level. In Lithuania, citizens were able to use coupons to buy public housing in which they resided. In Poland, where the government worries more about corporate governance, bidding will rely more heavily on investment funds than on individual citizens. What can the Czech and Slovak experience teach?
Pragmatism should be the guiding principle. Too often privatization programs stall on methodological debates. Instead, the Czechs and Slovaks were pragmatic, using whatever method would achieve their privatization objectives most quickly. The coupon scheme was just one more tool in the government’s tool-kit, which included a variety of more conventional tools as well.
Keep it simple. Besides ownership transformation, privatization is often asked to address regional development, unemployment, and fiscal problems. Under the first wave, the Czech and Slovak Governments used other policies to address these problems, relying on privatization only to create profitable private firms that would pay taxes and create jobs. The introduction of multiple objectives in the second wave of privatization in the Slovak Republic has been a major reason for delays.
Minimize government meddling in politically important firms. In most transition economies, there tend to be a small number of large insolvent firms, often in isolated locales, that almost inevitably get special treatment because of political importance. These firms are ill-suited to mass privatization and should probably be excluded. However, insolvency per se is not a good reason for exclusion since private owners are better than government at restructuring. Government even-handedness in the form of hard budget constraints is key for achieving improvements in enterprise performance.
Better information makes for better markets. The government made substantial information publicly available at the start of the process, which played an important role in determining share prices in the early bidding rounds. The special role of information becomes apparent when compared to some alternatives. In more centralized approaches, such as in Poland, public information is less important since the design of the program relies more on investment funds to be informed, to be diversified to protect citizens who own shares in funds, and to exercise governance. Where insiders (workers and managers) are major players in mass privatization schemes (such as in Russia), the information problem may be smaller (because insiders tend to already have considerable information about their firms), although the efficiency gains for stock markets that result from making information publicly available are not realized.
Sequential auctions allow bidders to learn. The use of a sequence of simultaneous bidding rounds both used and created information by forcing those with knowledge about enterprises’ true values to reveal it to others. Consider some alternatives. A single simultaneous bidding round that allowed people to make multiple bids (thereby revealing their demand curve) could have been used, but that would have favored those with access to privileged information and generated no intermediary information for other bidders. Similarly an auction that was not simultaneous, but was conducted individually for each enterprise would not have generated the relative price information that facilitated the emergence of a market equilibrium.
Privatization is always political. Like most reforms, there are winners and losers, and the program’s design must adapt to the political reality. In the Czech and Slovak Republics, there was a strong political commitment to an equitable process, there was an ability to overcome strong vested interests, and an administrative infrastructure and a citizenry existed with the capacity to implement a fairly sophisticated scheme. Conditions elsewhere may not be so favorable, meaning that political and administrative constraints will require simpler schemes, a greater need to placate powerful “insiders,” and perhaps less equity among beneficiaries.
For more details, see the author’s, “Making a Market: Mass Privatization in the Czech and Slovak Republics,” World Bank Policy Research Working Paper 1231, December 1993, and “Information and Price Determination under Mass Privatization,” World Bank Policy Research Working Paper 1305, June 1994.