Journal of Economic LiteratureVol. XXXIX (June 2001) pp. 321–389

Megginson and Netter: From State to MarketJournal of Economic Literature, Vol. XXXIX (June 2001)

From State to Market:A Survey of Empirical Studies

on Privatization

WILLIAM L. MEGGINSON and JEFFRY M. NETTER1

1. Introduction

THE POLITICAL AND economic policyof privatization, broadly defined as

the deliberate sale by a government of

state-owned enterprises (SOEs) or assetsto private economic agents, is now in useworldwide. Since its introduction byBritain’s Thatcher government in theearly 1980s to a then-skeptical public(that included many economists), privati-zation now appears to be accepted as alegitimate—often a core—tool of state-craft by governments of more than 100countries. Privatization is one of themost important elements of the continu-ing global phenomenon of the increasinguse of markets to allocate resources.

It is tempting to point to the spreadof privatization programs around theworld during the past two decades andconclude that the debate on the eco-nomic and political merits of govern-ment versus private ownership has beendecided. But such a conclusion isflawed, since 25 years ago proponentsof state ownership could just as easilyhave surveyed the postwar rise of state-owned enterprises and concluded thattheir model of economic organizationwas winning the intellectual battle withfree-market capitalism. Instead ofpointing to the spread of privatizationand calling it destiny, our goal is to as-sess the findings of empirical research onthe effects of privatization as a policy.Therefore, this paper surveys the rap-idly growing literature on privatization,

321

1 Megginson: University of Oklahoma. Netter:University of Georgia. This paper was developedwith financial support from the SBF Bourse deParis and the New York Stock Exchange, and theassistance of George Sofianos, Bill Tschirhart, andDidier Davidoff is gratefully acknowledged. Weappreciate comments received on this paper fromGeert Bekaert, Anthony Boardman, Bernardo Bor-tolotti, Narjess Boubakri, Jean-Claude Cosset,Kathryn Dewenter, Alexander Dyck, OlehHavrylyshn, Ivan Ivanov, Jonathan Karpoff, RankoJelic, Claude Laurin, Marc Lipson, Luis López-Calva, John McMillan (the editor), Sandra Sizer-Moore, Harold Mulherin, Rob Nash, John Nellis,David Newberry, David Parker, Enrico Perotti,Annette Poulsen, Ravi Ramamurti, Susan Rose-Ackerman, Nemat Shafik, Mary Shirley, MikeStegmoller, Aidan Vining and three anonymousreferees. We appreciate comments from partici-pants at the NYSE/Paris Bourse Global EquityMarkets conference (Paris, Dec. 1998), HarvardInstitute for International Development Privatiza-tion Workshop (June 2000), International Federa-tion of Stock Exchanges’ Third Global EmergingMarkets Conference (Istanbul, April 2000), WorldBank and/or IFC meetings, OECD conferences(Paris and Beijing), 1999 Conference on Privatiza-tion and the Kuwaiti Economy in the Next Cen-tury, 1998 Financial Management Associationmeeting, 1999 European Financial ManagementAssociation meeting, Fondazione ENI EnricoMattei (FFEM), Swiss Banking Institute andCredit Suisse, and seminars at City UniversityBusiness School (London), London Guildhall Uni-versity, and University of Oklahoma. All remainingerrors are the authors’ alone.

attempts to frame and answer the keyquestions this stream of research hasaddressed, and then describes some ofits lessons on the promise and perils ofselling state-owned assets. Throughoutthis survey, we adopt the perspective ofan advisor to a government policymakerwho is wrestling with the practicalproblems of whether and how to imple-ment a privatization program. The poli-cymaker asks “What does the researchliterature have to tell us about these as-pects of privatization as an economicpolicy?” We attempt to answer theseimportant questions.

This paper is organized as follows.Section 2 provides a brief historicaloverview of privatization. We examinethe impact that privatization programshave had in reversing SOE involvementin the economic life of developed anddeveloping countries. Section 3 brieflysurveys the recent theoretical and em-pirical research on the relative eco-nomic performance of state-owned andprivately owned firms. Section 4 detailsthe different types of transactions thatare labeled “privatization” in differentregions. We draw particular attention tothe structure and pricing selected forshare issue privatizations. We alsoevaluate the various forms of “voucher”or “mass” privatizations that have beenimplemented. This section also exam-ines whether less radical methods of im-proving the performance of SOEs, suchas deregulation and allowing greatercompetition (or more routine steps suchas using management performance con-tracts), can effectively substitute foroutright privatization. In section 5, weexamine the issue of whether, and byhow much, privatization programs haveactually improved the economic and fi-nancial performance of divested firms.Our discussion first evaluates privatiza-tion in industrialized and developingcountries, and then assesses privatiza-

tion’s overall impact in the transitioneconomies. Section 6 asks whether do-mestic and international investors whopurchase privatizing share offerings ex-perience positive initial and long-terminvestment returns, and section 7 evalu-ates the impact of privatization on thedevelopment of non-U.S. capital mar-kets over the past two decades. Finally,section 8 discusses how privatizationprograms have impacted the develop-ment of—and interest in—corporategovernance practices around the world.Section 9 concludes and summarizesour survey.

2. How Large Has Privatization’sImpact Been to Date?

Given the attention the press hasgiven to the global movement towardmarkets, especially the privatization ofstate-owned enterprises, some mightconclude that privatization has almostended the involvement of state-ownedenterprises in global economic activity.2

This is a significant overstatement. Tounderstand the impact of privatizationon the state’s role in different econo-mies, we must first briefly review thehistory behind both privatization and itsprecursor, nationalization.

Throughout history, there has been amixture of public (often including reli-gious institutions) and private ownershipof the means of production and com-merce. Robert Sobel (1999) writes thatstate ownership of the means of produc-tion, including mills and metal working,was common in the ancient Near East,while private ownership was more com-mon in trading and money lending. In

2 Throughout this paper, we will use the WorldBank’s definition of state-owned enterprises, asdescribed in World Bank (1995): “government-owned or government-controlled economic enti-ties that generate the bulk of their revenues fromselling goods and services.”

322 Journal of Economic Literature, Vol. XXXIX (June 2001)

ancient Greece, the government ownedthe land, forests, and mines, but con-tracted out the work to individuals andfirms. In the Ch’in dynasty of China,the government had monopolies on saltand iron. Sobel notes that in the RomanRepublic the “publicani (private indi-viduals and companies) fulfilled virtu-ally all of the state’s economic require-ments.” Dennis Rondinelli and MaxIacono (1996) note that by the time ofthe Industrial Revolution in the westernindustrialized societies and their colo-nies, the private sector was the most im-portant producer of commercial goodsand was also important in providingpublic goods and services. This pattern,with more government involvement insome countries and less in others, con-tinued into the twentieth century inwestern Europe and its colonies andformer colonies. In the United States,there was less government involvementthan in many other countries.

The Depression, World War II, andthe final breakup of colonial empirespushed government into a more activerole, including ownership of productionand provision of all types of goods andservices, in much of the world. In west-ern Europe, governments debated howdeeply involved the national govern-ment should be in regulating the na-tional economy and which industrialsectors should be reserved exclusivelyfor state ownership. Until MargaretThatcher’s conservative governmentcame to power in Great Britain in 1979,the answer to this debate in the UnitedKingdom and elsewhere was that thegovernment should at least own thetelecommunications and postal services,electric and gas utilities, and mostforms of non-road transportation (espe-cially airlines and railroads). Many poli-ticians also believed the state shouldcontrol certain “strategic” manufactur-ing industries, such as steel and defense

production. In many countries, state-owned banks were also given either mo-nopoly or protected positions, as dis-cussed in Rafael La Porta, FlorencioLópez-de-Silanes, and Andrei Shleifer(2000a).

Rondinelli and Iacono (1996) arguethat government ownership grew in thedeveloping world for slightly differentreasons, primarily that government own-ership was perceived as necessary topromote growth. In the post-colonialcountries of Asia, Africa, and LatinAmerica, governments sought rapidgrowth through heavy investment in physi-cal facilities. Another reason for govern-ment ownership, often through nation-alization, was a historical resentment ofthe foreigners who had owned many ofthe largest firms in these countries (seealso Roger Noll 2000).

Thus there had been tremendousgrowth in the use of SOEs throughoutmuch of the world, especially afterWorld War II, which in turn led to pri-vatizations several decades later.3 Mostpeople associate modern privatizationprograms with Thatcher’s government.However, the Adenauer government inthe Federal Republic of Germanylaunched the first large-scale, ideologi-cally motivated “denationalization” pro-gram of the postwar era. In 1961, theGerman government sold a majoritystake in Volkswagen in a public shareoffering heavily weighted in favor ofsmall investors.4 Four years later, the

3 The historical overview of postwar privatiza-tions is based on a longer historical discussion inMegginson, Robert Nash, and Matthias van Ran-denborgh (1994). Other discussions of the histori-cal evolution of privatization include TimothyJenkinson and Colin Mayer (1988), Shirley andJohn Nellis (1991), World Bank (1995), JosefBrada (1996), Paul Bennell (1997), and DanielYergin and Joseph Stanislaw (1998).

4 Using a broader definition of privatization—one that encompassed reactively changing the poli-cies of an immediate predecessor government—the Churchill government’s denationalization of

Megginson and Netter: From State to Market 323

government launched an even larger of-fering for shares in VEBA. Both offer-ings were initially received favorably,but the appeal of share ownership didnot survive the first cyclical downturnin stock prices, and the government wasforced to bail out many small sharehold-ers. It was almost twenty years beforeanother major western nation chose topursue privatization as a core economicor political policy.5

Although the Thatcher governmentmay not have been the first to launch alarge privatization program, it is with-out question the most important histori-cally. Privatization was not a major cam-paign theme for the Tories in 1979, butthe new conservative government em-braced the policy. Thatcher adopted thelabel “privatization,” which was origi-nally coined by Peter Drucker andwhich replaced the term “denationaliza-tion” (Yergin and Stanislaw 1998, p.114). Early sales were strenuously at-tacked by the Labour opposition, whichpromised that if it were reelected itwould renationalize divested firms suchas British Aerospace and Cable andWireless.6

It was not until the successful BritishTelecom initial public offering in No-vember 1984 that privatization becameestablished as a basic economic policyin the United Kingdom. A series of in-creasingly massive share issue privatiza-tions (SIPs) during the last half of the1980s and the early 1990s reduced therole of SOEs in the British economyto essentially nothing after the Toriesleft office in 1997, from more than 10percent of GDP eighteen years earlier.

We note that the objectives set forthe British privatization program by theConservatives were virtually the sameas those listed by the Adenauer govern-ment twenty years before—and almostevery government since. These goals,as described in Price Waterhouse(1989a,b), are to (1) raise revenue forthe state, (2) promote economic effi-ciency, (3) reduce government interfer-ence in the economy, (4) promote widershare ownership, (5) provide the oppor-tunity to introduce competition, and (6)subject SOEs to market discipline. Theother major objective mentioned by theThatcher and subsequent governmentswas to develop the national capital mar-ket.7 We note these goals can be con-flicting and we discuss the trade-offsfurther in the paper.

The perceived success of the Britishprivatization program helped persuademany other industrialized countries tobegin divesting SOEs through publicshare offerings. Jacques Chirac’s gov-ernment, which came to power inFrance in 1986, privatized 22 compa-nies (worth $12 billion) before beingousted in 1988. The returning socialistgovernment did not execute any furthersales, but neither did it renationalizethe divested firms. Beginning in 1993,

the British steel industry during the early 1950scould well be labeled the first “privatization.” Wethank David Parker for pointing this out to us.

5 Pan Yotopoulos (1989) describes and assessesthe Chilean programs, which began before theprogram in the U.K. The Pinochet government ofChile, which gained power after the ouster of Sal-vador Allende in 1973, attempted to privatize com-panies that the Allende government had national-ized. However, the process was poorly executedand required very little equity investment frompurchasers of assets being divested. Thus, many ofthese same firms were renationalized once Chileentered its debt and payments crisis in the early1980s. Chile’s second privatization program, whichwas launched in the mid-1980s and relied more onpublic share offerings than direct asset sales (inwhich the government often acted as creditor aswell as seller) was much more successful.

6 Ironically, a labor government partially priva-tized an SOE just before Thatcher came to power.In 1977, the Labour government sold a relativelysmall fraction of the government’s shares in Brit-ish Petroleum as a means of raising cash.

7 Kojo Menyah, Krishna Paudyal, and CharlesInganyete (1995) and Menyah and Paudyal (1996)have more detailed discussions of the goals of theBritish privatization program.

324 Journal of Economic Literature, Vol. XXXIX (June 2001)

the Balladur government launched anew and even larger French privatiza-tion program, which has continued un-der the Jospin administration. The So-cialists, in fact, launched the two largestFrench privatizations ever, the $7.1 bil-lion France Telecom initial public of-fering (IPO) in October 1997 and thesubsequent $10.5 billion seasonedFrance Telecom issue in November 1998.

Several other European governments,including those of Italy, Germany, and,most spectacularly, Spain, also launchedlarge privatization programs during the1990s. These programs typically reliedon public share offerings, and wereoften launched by avowedly socialistgovernments. Privatization spread tothe Pacific Rim, beginning in the late1980s. Japan has sold only a relativehandful of SOEs during the past fifteenyears (usually relying on SIPs), butmany of these have been truly enor-mous. The three Nippon Telegraph andTelephone share offerings executed be-tween February 1987 and October 1988raised almost $80 billion, and the $40billion NTT offer in November 1987 re-mains the largest single security offeringin history. Elsewhere in Asia, govern-ments have taken an opportunistic ap-proach to SOE divestment, selling piecesof large companies when market condi-tions are attractive, or when money isneeded to plug budget deficits. It is un-clear how the economic difficulties thatgripped the region during the late 1990swill impact privatizations in the future.

Two Asian countries deserve specialattention. These two countries are al-ready the world’s second and fifth larg-est economies on a purchasing-power-parity basis, and promise to becomeeven more important over time. ThePeople’s Republic of China launched amajor economic reform and liberaliza-tion program in the late-1970s that hastransformed the productivity of the Chi-

nese economy. While there have beennumerous small privatizations, therehave been relatively few outright salesof SOEs, thus the overall impact of pri-vatization has been limited. Though thegovernment recently (1999) reaffirmedits commitment to privatizing all butthe very largest state enterprises, thefact that Chinese SOEs are burdenedwith so many social welfare responsi-bilities suggests that it will be extraordi-narily difficult to implement a privatiza-tion program large enough to seriouslyundermine the state’s economic role(Cyril Lin 2000 ; Justin Lin, Fang Cai,and Zhou Li 1998; and Chong-en Bai,David Li, and Yijaiang Wang 1997). Theother special Asian case is India, whichadopted a major economic reform andliberalization program in 1991, after be-ing wedded to state-directed economicdevelopment for the first 44 years of itsindependence. India’s reform programshares two key features with China’s: itwas adopted in response to highly dis-appointing SOE performance (SumitMajumdar 1996), and privatization hasthus far not figured prominently in thereform agenda.

On the other hand, Latin America hastruly embraced privatization. Chile’sprogram is particularly important, bothbecause it was Latin America’s first andbecause the 1990 Telefonos de Chileprivatization, which used a large Ameri-can depository receipt (ADR) sharetranche targeted toward U.S. investors,opened the first important pathway fordeveloping countries to directly tapwestern capital markets.

Mexico’s program was both vast inscope and remarkably successful at re-ducing the state’s role in what had beenan interventionist economy. La Portaand López-de-Silanes (1999) report thatin 1982, Mexican SOEs produced 14percent of GDP, received net transfersand subsidies equal to 12.7 percent of

Megginson and Netter: From State to Market 325

GDP, and accounted for 38 percent offixed capital investment. By June 1992,the government had privatized 361 of itsroughly 1,200 SOEs, and the need forsubsidies had been virtually eliminated.

Several other countries in LatinAmerica have also executed large di-vestment programs (Pablo Gottret1999). For example, Bolivia’s innovative“capitalization” scheme has been widelyacclaimed. However, the most impor-tant program in the region is Brazil’s.Given the size of Brazil’s economy andits privatization program, and the factthat the Cardoso government was ableto sell several very large SOEs (CVRDin 1997 and Telebras in 1998) in spiteof significant political opposition, thiscountry’s program is likely to remainvery influential.

Privatization in sub-Saharan Africahas been something of a stealth eco-nomic policy. Few governments haveopenly adopted an explicit SOE divest-ment strategy, but Bennell (1997)shows that there has been substantiallymore privatization in the region than iscommonly believed. For example,Steven Jones, Megginson, Robert Nashand Netter (1999) show that Nigeria hasbeen one of the most frequent sellers ofSOEs, using public share offerings, al-though they were very small. The expe-rience of the African National Congressafter it came to power in South Africaalso shows the policy realities that gov-ernments with interventionist instinctsface in this new era. Though nationali-zation and redistribution of wealth havebeen central planks of ANC ideology fordecades, the Mandela and Mbeki gov-ernments have almost totally refrainedfrom nationalization and have even soldoff several SOEs (though use of theword “privatization” remains taboo).

The last major region to adopt priva-tization programs comprises the formerSoviet-bloc countries of central and

eastern Europe. These countries beganprivatizing SOEs as part of a broadereffort to transform themselves fromcommand to market economies. There-fore, they faced the most difficult chal-lenges and had the most restricted setof policy choices. After the collapse ofcommunism in 1989–91, all of thenewly elected governments of the re-gion were under pressure to createsomething resembling a market econ-omy as quickly as possible. However,political considerations essentially re-quired these governments to significantlylimit foreign purchases of divested assets.

Since the region had little financialsavings, these twin imperatives com-pelled many—though not all—govern-ments throughout the region to launch“mass privatization” programs. Theseprograms generally involved distrib-uting vouchers to the population, whichcitizens could then use to bid for sharesin companies being privatized. Theprograms resulted in a massive reduc-tion of state ownership and were ini-tially popular politically, but becameunpopular in many countries (especiallyRussia ) because of the largely correctperception that they were robbery bythe old elite and the new oligarchs. Thenet effects have been disappointing insome cases, but have varied widely. Wediscuss the empirical evidence onvoucher privatization in section 5.

Although different regions have em-braced privatization at varying speeds,governments have found the lure ofrevenue from sales of SOEs to be at-tractive—which is one reason the policyhas spread so rapidly. According to Pri-vatisation International (Henry Gibbon1998, 2000), the cumulative value ofproceeds raised by privatizing govern-ments exceeded $1 trillion sometimeduring the second half of 1999. As anadded benefit, this revenue has come togovernments without raising taxes or

326 Journal of Economic Literature, Vol. XXXIX (June 2001)

cutting other government services. An-nual proceeds grew steadily before peak-ing at over $160 billion in 1997. Sincethen, proceeds seem to have leveled offat an annual rate of about $140 billion.Figure 1 shows the annual revenuesgovernments have received from privati-zations from 1988 through 1999. LadanMahboobi (2000) reports similar figuresclassified by privatizations in OECDand non-OECD countries. He reportsthat since 1990 privatization in OECDcountries has raised over $600 billion,approximately two-thirds of global pri-vatization activity. Western Europe hasaccounted for over half of these pro-ceeds. Finally, Jeffry Davis, Rolando Os-sowski, Thomas Richardson, and StevenBarnett (2000) report for a sample oftransition and non-transition countriesthat privatization proceeds were an averageof one and three-quarters percent of GDP.

The historical discussion suggests thatstate ownership has been substantiallyreduced since 1979, and in most coun-tries this has in fact occurred. Using datafrom Eytan Sheshinski and Luis FelipeLópez-Calva (1999), figure 2 demonstrates

the role of state-owned enterprises inthe economies of high-income (industri-alized) countries has declined signifi-cantly, from about 8.5 percent of GDPin 1984 to less than 6 percent in1991. Data presented in James Schmitz(1996), Mahboobi (2000), and BernadoBortolotti, Marcella Fantini, andDomenico Siniscalco (1999a), as well asour own empirical work on share issueprivatizations suggests that the SOEshare of industrialized-country GDPhas continued to decline since 1991,and is now probably below 5 percent.

The low-income countries show aneven more dramatic reduction in stateownership. From a high point of almost16 percent of GDP, the average SOEshare of national output dropped tobarely 7 percent in 1995, and has prob-ably dropped to about 5 percent sincethen. The middle-income countries alsoexperienced significant reductions instate ownership during the 1990s. Sincethe upper- and lower-middle-incomegroups include the transition economiesof central and eastern Europe, this de-cline was expected, given the extremely

1988 1990 1992 1994 1996 1998

180

160

140

120

100

80

60

40

20

0

Figure 1. Annual Privatization Revenues for Divesting Governments, 1988–99

$US Billions

Source: Privatisation International.

Megginson and Netter: From State to Market 327

high beginning levels of state ownership.For example, Nemat Shafik (1995) reportsthat the Czechoslovakian governmentowned 98 percent of all property in 1989.

3. Why Have Governments EmbracedPrivatization?

3.1 Efficiency of State vs PrivateOwnership: Theory

Throughout history, scholars, includ-ing economists, have debated the roleof government in the economy.8 Among

economists, this debate now spans manyareas, including welfare economics,public choice, public finance, industrialorganization, law and economics, corpo-rate finance, and macroeconomics. Inthis section, we summarize some of theimportant theoretical issues that arisein the study of privatization and that areneeded to analyze the empirical evi-dence we review in the rest of the pa-per. We concentrate on empirical evi-dence because, as Jean-Jacques Laffontand Jean Tirole (1993) say after pre-senting their model analyzing trade-offsbetween government and private own-ership in promoting efficiency, “theoryalone is thus unlikely to be conclusivein this respect.” There are also severalexcellent articles that discuss the the-ory of privatization and review the lit-erature, including Anthony Boardmanand Aidan Vining (1989), John Vickers

8 For example, Friedrich von Hayek’s (1994)passionate critiques of the welfare state and col-lectivism, exemplified in the 1944 book The Roadto Serfdom, had a direct impact on policymakers indeveloping a motive for privatization. Yergin andStanislaw (1998, pp. 98–107) discuss how Hayek’swork was the intellectual basis for Keith Josephand then Thatcher and the Tory politicians whobegan the intellectual campaign against statism inthe U.K. that triggered the worldwide privatiza-tion movement.

18

16

14

12

10

8

6

4

2

0

Figure 2. SOE Share of GDP by State of National Development, 1979–96

Source: World Bank, as reported in Sheshinski and López-Calva (1999).

SOE as % of GDP

Low income

1995199019851980

Lower-middle income

Upper-middle income

High income

328 Journal of Economic Literature, Vol. XXXIX (June 2001)

and George Yarrow (1991), Shleifer(1998), Oleh Havrylyshyn and DonaldMcGettigan (2000), John Nellis (1999,2000), Sheshinski and López-Calva(1999), Simeon Djankov and PeterMurrell (2000a,b) and Shirley andPatrick Walsh (2000).

The economic theory of privatizationis a subset of the large literature on theeconomics of ownership and the rolefor government ownership (or regula-tion) of productive resources. An initialquestion to be asked is “what is theproper role of government?” Implicitly,we assume that the goal of governmentis to promote efficiency. Thus, we dis-cuss the efficiency implications of gov-ernment ownership and, more impor-tantly, the movement from governmentownership to privatization. To a largeextent we ignore the arguments regard-ing the importance of equitable con-cerns such as income distribution, be-cause they are beyond the scope of thisreview. The effects of privatization onproductive efficiency, or at least observ-able variables that are proxies for pro-ductive efficiency, is the focus of most ofthe empirical literature we review here.

The theoretical arguments for the ad-vantages of private ownership of themeans of production are based on afundamental theorem of welfare eco-nomics: Under strong assumptions, acompetitive equilibrium is pareto opti-mal. However, the assumptions includerequirements that there are no exter-nalities in production or consumption,that the product is not a public good,that the market is not monopolistic instructure, and that information costsare low. Thus, a theoretical argumentfor government intervention based onefficiency grounds rests on an argumentthat markets have failed in some way,one or more of these assumptions donot hold, and that the government canresolve the market failure.

Intellectual arguments for govern-ment intervention based on efficiencyconsiderations have been made in manyareas. Governments perceive the needto regulate (or own) natural monopoliesor other monopolies, intervene in thecase of externalities (such as regulatingpollution), and help provide publicgoods (such as providing national de-fense and education, or in areas wherethere is a public good aspect to provid-ing information). The arguments forgovernment intervention become morecomplicated when they extend to distri-butional concerns. For example, someargue that the role of government is toact as a “welfare state” (A. Briggs 1961),using state intervention in the marketeconomy to modify the actions of themarket.9 Thus, the arguments for stateownership or control rest on some ac-tual or perceived market failure, andcountries have often responded to mar-ket failure with state ownership. Privati-zation, in turn, is a response to thefailings of state ownership. Sometheoretical arguments that have arisen inthe privatization debate are discussednext.

3.1.1. The impact of privatization de-pends on the degree of market failure.As noted above, welfare theory (ignor-ing the theory of second best) arguesthat privatization tends to have thegreatest positive impact in cases wherethe role for government in lesseningmarket failure is the weakest, i.e., forSOEs in competitive markets or mar-kets that can readily become competi-tive. Sheshinski and López-Calva(1999), in summarizing the theoreti-cal literature, argue that there shouldbe “. . . important efficiency gainsfrom changes to private ownership in

9 I. Gough (1989) notes that Briggs (1961)claims that Archbishop Temple first used the termin wartime Britain to differentiate Britain from the“warfare” state of Nazi Germany.

Megginson and Netter: From State to Market 329

competitive structures.” In fact, the effectsof competition can be so strong thatSOEs, in an increasingly global environ-ment, may be forced to respond to pres-sures that maximize productive effi-ciency without the ownership change ofprivatization. (Shirley and Walsh 2000provide additional discussion of the ef-fects of competition on the privatizationdecision.)

In contrast, the justification for priva-tization is less compelling in marketsfor public goods and natural monopo-lies where competitive considerationsare weaker. However, Shleifer (1998)and others have argued that even inthose markets, government-owned firmsare rarely the appropriate solution, formany of the reasons discussed below.

3.1.2. Contracting ability impacts theefficiency of state and private owner-ship. Government ownership of firmsresults in problems in defining the goalsof the firm. While the shareholder-wealth-maximizing model of corporateorganization is becoming increasinglydominant in part because of the advan-tages of having a well-defined corporategoal (see Henry Hansmann and ReinierKraakman 2000), governments have otherobjectives than profit or shareholder-wealth maximization. Further, these ob-jectives can change from one adminis-tration to the next. Government’s in-ability to credibly commit to a policycan significantly reduce the efficiencyof an SOE’s operations and governance.Even if the government does attempt tomaximize social welfare, for example,welfare is a difficult thing to measureand use in guiding policy.10 In addition,

the government’s goals can be inconsis-tent with efficiency and maximizingsocial welfare, or even malevolent(see Laffont and Tirole 1993; Shleifer1998).

In addition, even if the government andthe nation’s citizens agree that profitmaximization is the goal of the firm, itis difficult to write complete contractsthat adequately tie managers’ incentivesto that goal. Shleifer (1998) argues thatthe owners of public firms (the nation’scitizens) are less able to write completecontracts with their managers becauseof diffuse ownership, making it difficultto tie the managers’ incentives to thereturns from their decisions. This is asubset of the broader arguments, basedon property rights and agency costs,that there will be differences in perfor-mance between government and pri-vately held firms because there is abroader range of monitoring devicesunder private ownership.11

3.1.3. Ownership structure affects theease with which government can inter-vene in firm operations. Governmentscan intervene in the operations of anyfirm, either public or private. However,the government’s transaction costs ofintervening in production arrangementsand other decisions of the firm aregreater when firms are privately owned.Thus, to the extent that government in-tervention has greater costs than benefits,private ownership is preferred to publicownership (see David Sappington andJoseph Stiglitz 1987).

3.1.4. A major source of inefficiency inpublic firms stems from less-prosperousfirms being allowed to rely on the

10 Stiglitz (1998) provides an insightful analysis,based on personal experience, of the difficultygovernments face in implementing pareto-efficient improvements due to information costsand the problems of commitment and dynamicbargaining. These arguments apply to both gov-ernment regulation (the main case Stiglitz ana-lyzes) and to state ownership.

11 Armen Alchain (1977, p. 36) notes, “behaviorunder [public and private] ownership is different,not because the objectives sought by organizationsunder each form are different, but, instead, be-cause even with the same explicit organizationgoals, the costs-rewards system impinging on theemployees and the ‘owners’ of the organization aredifferent.”

330 Journal of Economic Literature, Vol. XXXIX (June 2001)

government for funding, leading to“soft” budget constraints. The state isunlikely to allow a large SOE to facebankruptcy. Thus the discipline en-forced on private firms by capital mar-kets and the threat of financial distressis less important for state-owned firms.János Kornai (1988, 1993, 2000), EricBerglof and Gérard Roland (1998), andRoman Frydman, Cheryl Gray, MarekHessel, and Andrzej Rapaczynski (2000)all suggest that soft budget constraintswere a major source of inefficiency incommunist firms. They also note thatsupposedly “hard” budget constraintsimposed on SOEs by government arenot very effective either.

3.1.5. Privatization can impact effi-ciency through its effect on governmentfiscal conditions. As noted in section 1,governments have raised huge amountsof money by selling SOEs. Such saleshave helped reduce the fiscal deficit inmany countries. Though important, ex-amining the efficiency effects of reduc-ing government deficits is beyond thescope of this paper. Davis et al. (2000)review the evidence on the macro-economic effects of privatization, dis-cuss the difficulties of using macro-economic privatization data and reportsome evidence on the effects fromeighteen developing countries. They findevidence that the proceeds from privati-zation are saved by governments and notused to increase government spending.

3.1.6. At a macroeconomic level, pri-vatization can help develop product andsecurity markets and institutions. Oneimportant motivation for privatization isto help develop factor and product mar-kets, as well as security markets. As dis-cussed above, welfare economics arguesthat efficiency is achieved through com-petitive markets. Thus, to the extentthat privatization promotes competi-tion, privatization can have importantefficiency effects. Inevitably, the effec-

tiveness of privatization programs andmarkets themselves are simultaneouslydetermined. It has been clear in thetransition economies that the success ofprivatization depends on the strength ofthe markets within the economies, andvice versa. Thus, the impact of privati-zation will differ across countries de-pending on the strength of the existingprivate sector. Similarly, evidence sug-gests that the effectiveness of privatiza-tion depends on institutional factorssuch as the protection of investors. How-ever, privatization can also stimulatethe development of institutions thatimprove market operations.

3.2 Summary of Privatization Theory

Theoretical work that examines priva-tization offers many reasons why, evenin the case of market failure, state own-ership has important weaknesses. AsShleifer (1998) sums up much of the lit-erature, “. . . a good government thatwants to further ‘social goals’ would rarelyown producers to meet its objectives.”A question for the post-privatizationworld is the role of the public sector inthe economy and in the regulation offirms. The alternative to state owner-ship is rarely purely private, unregu-lated firms. State ownership is only oneform of the continuum of governancestructures that reflect the level of stateregulation of public and privatelyowned firms (Laffont and Tirole 1993).Many of the theoretical arguments forprivatization are based on the premisethat the harmful effects of state inter-vention have a greater impact understate ownership than under state regu-lation, not that the harmful effects canbe eliminated through privatization.However, in this paper we leave to oth-ers the continuing debate on the properrole of regulation in a market-orientedeconomy. Instead, we analyze recentempirical literature examining the

Megginson and Netter: From State to Market 331

relative effectiveness of state versusprivate ownership.12

3.3 Efficiency of State vs PrivateOwnership: Empirical Evidence

Comparing the performance ofgovernment-owned to privately ownedfirms is one method through which theimpact of government ownership onfirm performance can be analyzed.13 Insection 5 we present a more completediscussion of the potential problems inall empirical work in this area, which in-cludes lack of data and bad data, omit-ted variables, endogeneity, and selectionbias. There are two methodological dif-ficulties that are especially pronouncedin attempts to isolate the impact ofownership on performance. First, incomparing SOEs to privately ownedfirms, it is difficult, if not impossible, todetermine the appropriate set of com-parison firms or benchmarks, especiallyin developing economies with limitedprivate sectors. Second, there are gen-erally fundamental reasons why certainfirms are government owned and othersare privately owned, including the de-

gree of perceived market failure withinthe particular industry. These factorsthat determine whether the firm is pub-licly or privately owned likely also havesignificant effects on performance.Thus, it is difficult to evaluate the ef-fects of government ownership wherethe ownership structure is itself en-dogenous to the system that includesboth political and performance goals.Despite these problems, researchershave compared SOE and private firmperformance in several cases with somesuccess. We summarize the papersincluded here in table 1.

Given the above noted limitations, Is-sac Ehrlich, George Gallais-Hamonno,Zhiqiang Liu, and Randall Lutter(1994) provide good evidence on pro-ductivity differences between state-owned and privately owned firms. Theyuse a sample of 23 comparable interna-tional airlines of different (and in somecases changing) ownership categoriesover the period 1973–83 for which theyare able to obtain good and comparablecost, output, and ownership data. Theydevelop a model of endogenous, firm-specific productivity growth as a func-tion of firm-specific capital and use themodel as a basis for their fixed-effectsregressions estimating a cost function ina simultaneous framework with input-demand equations. They argue thatthey are able to separate the impact ofownership changes on short-term levelsof productivity changes from the long-term effects on the rate of productivitygrowth, improving on earlier studiesthat concentrated on static rather thandynamic effects of, and changes in,state ownership. Further, they suggestthey are able to isolate the effects ofownership from other factors impact-ing the productivity growth rate, includ-ing market conditions and exogenoustechnical changes.

Ehrlich et al. (1994) find a significant

12 The opinions of policymakers throughout theworld have been moving closer to those expressedby Ronald Coase in his classic 1960 article, “TheProblem of Social Cost.” In analyzing market fail-ure, Coase says, “All solutions have costs, andthere is no reason to suppose that governmentalregulation is called for simply because the prob-lem is not handled well by the market or the firm.”James Brickley, Clifford Smith, and Jerold Zim-merman (2001, p. 54), in a more recent analysis,say markets have worked better because, “First,the price system motivates better use of knowl-edge and information in economic decisions. Sec-ond, it provides stronger incentives for individualsto make productive decisions.”

13 A related literature that we do not review ana-lyzes the relative performance of nonprofit firmsand for-profit firms. James Brickley and R.Lawrence Van Horn (2000), in an analysis of largehospitals, argue that the evidence suggests there islittle distinction between the behavior of nonprofitand for-profit hospitals. Their results suggest thesimilarities in behavior are due to the effects ofcompetition and not identical objective functionsof the managers.

332 Journal of Economic Literature, Vol. XXXIX (June 2001)

TABLE 1RECENT EMPIRICAL STUDIES ON PUBLIC VS PRIVATE OWNERSHIP

Study Sample description, study period, and methodology Summary of findings and conclusions

Boardmanand Vining1989

Examines economic performance of 500 largest non-US firms in 1983, classified by ownership structureas SOE, private, or mixed (ME). Employs 4profitability ratios and 2 measures of X-efficiency.

SOEs and MEs are significantly less profitable andproductive than private firms. MEs are no more prof-itable than pure SOEs—so full private ownership is re-quired to gain efficiency.

Vining andBoardman1992

Asks whether ownership “matters” in determiningefficiency of SOEs, or if only the degree of com-petition is important. Estimates performancemodel using 1986 data from 500 largest nonfinan-cial Canadian firms, including 12 SOEs and 93MEs.

After controlling for size, market share and other fac-tors, private firms are significantly more profitableand efficient than MEs and SOEs, though now findthat MEs outperform SOEs. Thus, ownership has aneffect separable from competition alone.

Pinto, Belka,and Krajewski1993

Tests whether privatization is required to improveperformance of SOEs by examining how Polish statesector responded in the 3 years after “Big Bang” re-forms of Jan. 1990, which liberalized prices, tight-ened fiscal/monetary policy and introduced competi-tion, without privatization.

Significant performance improvement due to macro-economic stabilization package, even without priva-tization; mostly due to hard budget constraints, tightbank lending policies, enhanced credibility of govern-ment’s “no bailout” pledge.

Ehrlich,Gallais-Hamonno,Liu, Lutter1994

Examines impact of state ownership on long-run rateof productivity growth and/or cost decline for 23 in-ternational airlines during 1973–83.

State ownership can lower long-run annual rate ofproductivity growth by 1.6–2.0% and rate of unit costby 1.7–1.9%. Ownership effects not affected by de-gree of competition.

Majumdar1996

Using industry-level survey data, compares perfor-mance of SOEs, MEs, and private Indian firms for1973–89. SOEs and MEs account for 37% of employ-ment and 66% of capital investment in India in 1989.

Documents efficiency scores averaging 0.975 for pri-vate firms, significantly higher than averages of 0.912for MEs and 0.638 for SOEs. State sector efficiencyimproves during “efficiency drives” but declinesafterwards.

Kole andMulherin 1997

Tests whether postwar performance of 17 firms partlyowned by US government due to seizure of “enemy”property during WWII differs significantly from per-formance of private US firms.

Though these firms experience abnormally high turn-over among boards of directors, manager tenure isstable, and SOE performance is not significantly dif-ferent from private firms.

Dewenter andMalatesta2001

Tests whether profitability, labor intensity, and debtlevels of SOEs listed among 500 largest non-USfirms in 1975, 1985, and 1995 differ from privatefirms on same lists.

After controlling for business cycles, finds privatefirms significantly (often dramatically) more profitable,have significantly less debt, and less labor intensiveproduction processes than SOEs.

LaPorta,Lopez-de-Silanes, andShleifer 2000a

Using data from 92 countries, examines whetherstate ownership of banks impacts financial system de-velopment and growth rates of economy and produc-tivity.

Extensive state ownership, especially in poorest coun-tries, retards financial system development and re-stricts economic growth rates, mostly due to impacton productivity.

Tian 2000 Studies relation between state shareholding and firmperformance of 825 publicly traded Chinese firms in1998. 413 had some government ownership, 312 hadnone.

Performance of “private” enterprises significantly su-perior to “mixed” enterprises. Corporate value gen-erally declines with state ownership, then increasesafter state share passes 45%.

Karpoff 2001 Examines 35 government financed and 57 privatelyfunded expeditions to the Arctic from 1819–1909.

Private expeditions performed better using severalmeasures of performance. More major discoverieswere made by private expeditions; most tragedies oc-curred on government-sponsored expeditions. Ro-bust results in regressions explaining expeditionoutcomes.

link between ownership and firm-specificrates of productivity growth. Their re-sults suggest that private ownershipleads to higher rates of productivitygrowth and declining costs in the longrun, and these differences are not af-fected by the degree of market compe-tition or regulation. Their estimatessuggest that the short-run effects ofchanges from state to private ownershipon productivity and costs are ambigu-ous, providing a possible explanationfor some of the anomalous results instudies. However, their point estimatesindicate that the change from completestate to private ownership in the longrun would increase productivity growthby 1.6 to 2 percent a year, while costswould decline by 1.7 to 1.9 percent.Their empirics also suggest that a par-tial change from state to private owner-ship has little effect on long-run pro-ductivity growth—the benefits arebased on complete privatization of thefirm.

This paper has advantages over muchof the other work in the area due to thegood data, as well as guidance from awell-developed literature in estimatingthe determinants of productivity. Theauthors perform some of the more so-phisticated econometric analysis of pa-pers in this area. For example, they rep-licate their results with a subset of firmsthat did not experience any within-firmchanges in ownership, enabling theauthors to be sure that their time-ownership interaction term capturesonly between-firm variations in owner-ship. Ehrlich et al. also perform variousother robustness checks using differentspecifications and subsamples, as wellas controlling for the special charac-teristics of their sample period (oilprice shocks and deregulation in theUnited States), and find that their re-sults are robust. Finally, they considerthe potential for simultaneity effects

between ownership and productivity,and find that causality goes from owner-ship to productivity, and not vice versa.The weakness in the work is that it isbased on one industry with relativelyold data. The authors also note thatthey make the implicit assumption thatall firms are cost minimizing, but ifstate-owned enterprises have other ob-jectives, it is difficult to interpret themeaning of differences in costs.

Sumit Majumdar (1996) examinesdifferences in efficiency betweengovernment-owned, mixed, and private-sector firms in India. He finds supportfor the superior efficiency of privateand mixed-sector firms over SOEs.Using aggregate, industry-level surveydata, Majumdar finds that SOEs ownedby the central and state governmentshave average efficiency scores of 0.658and 0.638, respectively, over the period1973–89. Mixed enterprises score 0.92,and private enterprises score 0.975. Aconcern with Majumdar’s study is thatthe aggregated nature of the data, alongwith problems arising from the relianceon survey data, limits his ability to iden-tify any specific areas where privateversus state ownership works best, andwhether there are simultaneity and se-lection bias problems in trying to esti-mate the effects of ownership and pro-ductivity. In addition, he can providelittle insight into the reasons for theefficiency differences between thesectors.

George Tian (2000) offers anothercountry-specific study. He examines 825companies listed on the Shanghai StockExchange, with 513 mixed-ownershipfirms and 312 private firms. He finds thatprivate firms perform better than mixedownership firms. In addition, he exam-ines the valuation of the companies andfinds that corporate value with smallgovernment shareholdings decreaseswith the fraction of state shareholding

334 Journal of Economic Literature, Vol. XXXIX (June 2001)

but rises when the government is alarge shareholder.

Another approach to studying the ef-fects of government ownership on effi-ciency relies on a multi-industry, multi-national, time-series methodology. Whilecross-sectional time series studies sufferfrom methodological problems we dis-cuss later, they are able to capture dif-ferences that are not apparent in single-country or single-industry studies. Aninfluential paper taking this approach isAnthony Boardman and Aidan R. Vining(1989) who examine the economic per-formance of the 500 largest non-U.S.industrial firms in 1983. Using fourprofitability ratios and two measures ofX-efficiency, they show that state-owned and mixed (state and private)ownership enterprises are significantlyless profitable and productive than areprivately owned firms. They also findthat mixed enterprises are no moreprofitable than SOEs, suggesting thatfull private control, not just partial own-ership, is essential to achieving perfor-mance improvement. In a later study,Vining and Boardman (1992) use a sam-ple of Canadian firms to re-examine thestate versus private ownership question.Their results are qualitatively similar totheir earlier findings. In addition, theCanadian study finds that mixed enter-prises are more profitable than SOEs,though they fall far short of private-firmlevels.

Kathryn Dewenter and Paul Malatesta(2001) follow the general approach ofBoardman and Vining (1989) using morerecent data. They test whether the prof-itability, labor intensity, and debt levelsof SOEs in the 500 largest internationalcompanies, as reported in Fortune for1975, 1985, and 1995, differ from pri-vately owned firms in the same samples.Their data have 1,369 total firm years,of which 147 represent government-owned firms. Since Fortune excluded

U.S. firms until 1995, the data aremainly international. After controllingfor firm size, location, industry, andbusiness-cycle effects, Dewenter andMalatesta find robust evidence that pri-vate companies are significantly (oftendramatically) more profitable thanSOEs, and also have lower levels of in-debtedness and fewer labor-intensiveproduction processes than do theirstate-owned counterparts.

Finally, Frydman, Gray, Hessel, andRapaczynski (1999) compare the perfor-mance of privatized and state firms inthe transition economies of CentralEurope, and explicitly try to control forselection bias.14 Using survey data for506 midsize manufacturing firms in theCzech Republic, Hungary, and Polandin 1994, they compare four measures offirm performance—sales revenues, em-ployment, labor productivity (revenueper employee) and material costs perunit of revenue. They compare the pri-vatized group to the nonprivatizedgroup with panel data, controlling forpotential pre-privatization differencesbetween the two groups. Frydman,Gray, Hessel, and Rapaczynski find thatthe average effect of privatization isthat it works—privatized firms performbetter than the state owned firms. How-ever, the performance improvement isconcentrated in revenue improvement(not cost reduction) in firms privatizedto outside owners.

Frydman et al. (1999) make two im-portant contributions. First, they showthat while privatization improves per-formance, the effect is limited to cer-tain measures of performance and caseswhere the SOE is sold to outside owners.Second, they attempt to control for theeffects of selection bias in examining

14 Frydman et al. also compare the performanceof the privatized firms to that of the firms whenthey were SOEs. Thus, we also discuss the paperin section 5 and it is summarized in table 5.

Megginson and Netter: From State to Market 335

the effects of privatization in severalways. They use a fixed effects model tocontrol for selection bias caused by un-observed firm characteristics correlatedwith performance outcomes that arefixed over time. Further, they contrastthe performance of firms privatized inone period with those privatized in an-other for two different time periods tocompare the privatized firms with howthey would have performed withoutprivatization. Finally, to control par-tially for the possibility that betterfirms are selected for privatization, theycontrast the pre-privatization perfor-mance of managerially controlled firmswith those controlled by other owners.Thus, the paper does an excellentjob of controlling for potential biases,though it necessarily depends on surveydata.

We conclude this section with twostudies that use unique situations toanalyze the effects of government ver-sus private ownership. Stacey Kole andJ. Harold Mulherin (1997) set out to an-swer the basic question in the publicversus private debate as posed by SamPeltzman (1971), “If a privately ownedfirm is socialized, and nothing else hap-pens, how will the ownership alone af-fect the firm’s behavior?” Kole andMulherin study seventeen firms withsignificant German or Japanese owner-ship when the United States enteredWorld War II. The U.S. government as-sumed ownership of the foreign stock inthese firms and ended up holding be-tween 35 and 100 percent of the com-mon stock for up to 23 years during andafter World War II. Kole and Mulherinfind industry controls for five firms,comprising 61 percent of the bookvalue of the seventeen firms, and com-pare the performance of the government-owned firms. They find no significantdifference between the performance oftheir sample with the private-sector

firms and state “the preceding resultsstand in contrast to the typical resultsregarding the inefficiency of govern-ment enterprise.” The authors arguethat the fact that these firms were oper-ating in competitive industries forcedthem to operate efficiently.

The Kole and Mulherin (1997) re-sults are evidence that in a competitiveenvironment, where the governmenthas no agenda other than as a passiveinvestor, factors other than ownershipdetermine firm performance. Many ofthe firms were involved in the war ef-fort, so the government had an incen-tive to run them efficiently. In addition,all the firms were eventually repriva-tized, so the government was also con-cerned with running them efficiently tomaximize the later sale value. Kole andMulherin admit that their sample andthe period they study is novel, limitingits generality. Further, their results arebased on only five firms. Still, their find-ings do illustrate the importance of factorsother than ownership in determiningfirm performance.

In a paper featuring a very interest-ing natural experiment, Jonathan Kar-poff (2001) studies a comprehensivesample of 35 government-funded and57 privately-funded expeditions to theArctic from 1818 to 1909 seeking to lo-cate and navigate a northwest passage,discover the North Pole, and makeother discoveries in arctic regions. Kar-poff finds that the private expeditionsperformed better using several measuresof performance. He shows most majorarctic discoveries were made by privateexpeditions, while most tragedies (lostships and lives) were on publiclyfunded expeditions. He notes the factthat the public expeditions had greaterlosses could mean the public expedi-tions took greater risks, but then thepublic expeditions would have had agreater share of discoveries, which did

336 Journal of Economic Literature, Vol. XXXIX (June 2001)

not occur. He also estimates regressionsexplaining outcomes in several ways(crew deaths, ships lost, tonnage ofships lost, incidence of scurvy, level ofexpedition accomplishment), control-ling for exploratory objectives sought,country of origin, the leader’s previousarctic experience, or the decade inwhich the expedition occurred. In es-sentially every regression, the dummyvariable for private expedition is signifi-cant with a sign indicating that the pri-vate expedition performed better. Kar-poff concludes that the incentives werebetter aligned in the private expeditions,leading to systematic differences in theways public and private expeditions wereorganized. While the uniqueness of thesample limits its generality, he providesan interesting illustration of the impactof ownership on the performance of anorganization.15

3.4 Policy Alternatives to Privatization

As discussed earlier, some argue thatcompetition and deregulation are moreimportant than privatization or gover-nance changes in improving perfor-mance of firms (George Yarrow 1986;John Kay and D. J. Thompson 1986;Matthew Bishop and Kay 1989; JohnVickers and Yarrow 1991; FranklinAllen and Douglas Gale 1999).

Others maintain that privatization isnecessary for significant performanceimprovements (Vining and Boardman1992; Maxim Boycko, Shleifer, andRobert Vishny 1994, 1996a,b; John Nel-lis 1994; Josef Brada 1996; and Shleifer1998). Although much of this debate isoutside the scope of this paper, thereare a few empirical studies that examinecountries where economic reform has

been implemented instead of, or priorto, full privatization.16

Brian Pinto, Merek Belka, and StefanKrajewski (1993) examine the way inwhich the Polish state sector respondedin the three years following Poland’s“Big Bang” reforms of January 1990.These reforms deregulated prices, in-troduced foreign competition to manyindustries, and signaled that tightmonetary and fiscal policies would bepursued. However, the Polish govern-ment did not immediately launch alarge-scale privatization program. Theauthors document significant perfor-mance improvements on the part ofmost manufacturing firms. They con-clude that these improvements weredue to the imposition of hard budgetconstraints reinforced by tighter banklending behavior, consistency in thegovernment’s “no bailout signal,” im-port competition, and reputationalconcerns of SOE managers.

The use of incentive contracts formanagement and workers is potentiallythe best way to improve performance inSOEs (Leroy Jones 1991). The WorldBank endorsed these contracts in the1980s. China has undergone widespreadeconomic reform with minimal privati-zation through the use of these incentivecontracts and offers a natural setting inwhich to study their impact.

Theodore Groves, Yongmiao Hong,John McMillan, and Barry Naughton(1995) discuss the ways incentives wereadded to the Chinese managerial labormarket by the late 1980s, includingreplacement after poor performanceand linking managerial pay to profits.Further, managers were selected byauctions, where the auction process

15 Kelly Olds (1994) also uses data from the1800s to show that after the privatization of thetax-supported Congregationalist churches in NewEngland, demand for preachers and church mem-bership rose dramatically.

16 Majumdar (1996) also suggests that reformcan improve SOE performance by showing thatthe gap between the private and public firms’ per-formance partly closes during those periods whengovernments are pushing reform agendas.

Megginson and Netter: From State to Market 337

revealed information about the manag-ers that in a market economy couldhave come from observations of theirperformance. Groves, Hong, McMillan,and Naughton (1994) show that after1978, when Chinese firms were givenmore autonomy and allowed to retainmore profits and to increase workers’incentives through bonuses and differingwork contracts, there were increasesin workers’ incomes (though not ofmanagers’) and in investment in thefirms.

Wei Li (1997) documents marked im-provements in the marginal and totalfactor productivity of 272 ChineseSOEs over the period 1980–89 as a re-sult of economic reforms in China, in-cluding the increased use of incentives.He finds evidence of substantial in-creases in productivity over the reformperiod, much of which can be attrib-uted to the reform. In addition, his evi-dence suggests that 87 percent of thegrowth in productivity was due to im-proved incentives and compensation. Linotes, however, the potential for selec-tion bias in his study both in the firmsselected for the survey and in theresponses to the survey.

Shirley and Lixin Xu (1998) come tothe opposite conclusion concerning theability of incentive contracts to improvefirm performance. They analyze the ef-fects of these contracts in twelve mo-nopoly SOEs, and find that the incentivecontracts have no effect on profitabilityor labor productivity; they also findsome evidence of negative effects ongrowth in total factor productivity. Theyattribute the failure of the contracts tothe inability of governments to followthrough on promised actions and the in-ability of supervisory agencies to negoti-ate and monitor the contracts effec-tively. It must be noted, however, thatthe study is based on a small sample,limiting the ability to draw conclusions,

especially in light of the evidence fromthe studies of Chinese firms.

The evidence from China suggeststhat enterprise restructuring, concen-trating on improving the allocation ofproperty rights and incentives can yieldlarge benefits even without privatiza-tion.17 Naturally, this begs the questionwhether economic reform coupled withprivatization could lead to even greaterperformance improvements. Unfortu-nately, this is little evidence on thisquestion and it would be very difficultto develop such evidence. Note also thatthe evidence on the benefits of reformwithout privatization comes primarilyfrom one country where country-specificfactors may play an important but un-identified role. One thing we can say isthat, as we note later in the paper, theevidence demonstrating the benefits ofprivatization is weakest for countries ineastern Europe, where privatization wasimplemented rapidly. This may suggestthat privatization should have pro-ceeded along a more gradual path. Weaddress that question later on.

4. How Do Countries Privatize?

A key decision to be made by the pri-vatizing government is on the methodof transferring the state-owned asset toprivate ownership. This decision is diffi-cult because, in addition to the eco-nomic factors such as valuing the assets,privatizations are generally part of anongoing, highly politicized process.Some of the factors that influence theprivatization method include: (1) thehistory of the asset’s ownership, (2)the financial and competitive positionof the SOE, (3) the government’s ideo-logical view of markets and regulation,

17 This is consistent with the findings of Brickleyand Van Horn (2000) that the managers of non-profit hospitals face similar incentives to the man-agers of for-profit hospitals and behave in a similarmanner.

338 Journal of Economic Literature, Vol. XXXIX (June 2001)

(4) the past, present, and potential fu-ture regulatory structure in the country,(5) the need to pay off important inter-est groups in the privatization, (6) thegovernment’s ability to credibly commititself to respect investors’ property rightsafter divestiture, (7) the capital marketconditions and existing institutionalframework for corporate governance inthe country, (8) the sophistication ofpotential investors, and (9) the govern-ment’s willingness to let foreigners owndivested assets.

The complexity of goals means thatcountries have used various methods forprivatizing different types of assets. Al-though financial economists havelearned much about selling assets inwell-developed capital markets, we stillhave a limited understanding of the de-terminants and implications of the pri-vatization method for state-owned as-sets. Theoreticians have modeled someaspects of the privatization process, butto be tractable, their models must ig-nore important factors. Empirical evi-dence on the determinants of privatiza-tion is also limited by the complexity ofthe goals of the privatization process.

4.1 Methods of Privatization

Brada (1996) presents an excellenttaxonomy of privatization methods. Al-though the context of his paper is cen-tral and eastern Europe, his classifica-tion of four principal divestmentmethods is quite general. In addition,he provides a review of the successesand failures of each of these general ap-proaches in central and eastern Europe.Of course, there are many variationswithin each of his categories, and heshows that many privatizations usecombinations of different methods.

Brada’s first category is privatizationthrough restitution. This method is ap-propriate when land or other easilyidentifiable property that was expropri-

ated in years past can be returned toeither the original owners or to their heirs.This method is rarely observed outsideeastern Europe, though it has been im-portant there. For example, Brada (1996)reports that up to 10 percent of thevalue of state property in the Czech Re-public consisted of restitution claims. Themajor difficulty with this method is thatthe records needed to prove ownershipare often inadequate or conflicting.

The second method is privatizationthrough sale of state property, where agovernment trades its ownership claimfor an explicit cash payment. This cate-gory takes two important forms. Thefirst is direct sales (or asset sales) ofstate-owned enterprises (or some partsthereof) to an individual, an existingcorporation, or a group of investors.The second form is share issue privati-zations (SIPs), in which some or all of agovernment’s stake in an SOE is sold toinvestors through a public share offer-ing. These are similar to IPOs in theprivate sector, but where private IPOsare structured primarily to raise reve-nue, SIPs are structured to raise moneyand to respond to some of the politicalfactors mentioned earlier.

Brada’s third category is mass orvoucher privatization, whereby eligiblecitizens can use vouchers that are dis-tributed free or at nominal cost to bidfor stakes in SOEs or other assets. Thismethod has been used only in the tran-sition economies of central and easternEurope, where it has brought aboutfundamental changes in the ownershipof business assets in those countries, al-though it has not always changed effec-tive control. Longer descriptions of theissues that these governments have con-fronted when designing voucher privati-zation programs are provided in MorrisBornstein (1994, 1999), Melinda Alex-androwicz (1994), Bernard Drum (1994)and Shafik (1995).

Megginson and Netter: From State to Market 339

The final method is privatizationfrom below, through the startup of newprivate businesses in formerly socialistcountries. Havrylyshyn and McGettigan(2000) stress the importance of thistype of economic growth in the transi-tion economies. Although privatizationfrom below has progressed rapidly inmany regions (including China, thetransition economies of central andeastern Europe, Latin America, andsub-Saharan Africa), a survey of thisphenomenon is beyond the scope of ourpaper.

There are many other methods be-sides the four described above that gov-ernments can use to increase private-sector participation. For example, theterm “privatization” in the United Statesmeans something different from any ofthese strategies. As López-de-Silanes,Shleifer, and Vishny (1997) show, the pri-vatization debate in the United States re-fers to the choice between provision ofgoods and services by (state and local)government employees and the contract-ing out of that production to privatefirms. Their empirical study finds thatthe more binding are state fiscal con-straints and the less powerful are public-sector unions, the greater the likelihoodof privatization.

4.2 The Choice of Sale Method

Henry Gibbon (1997) provides one ofthe most helpful delineations of the deci-sions facing a government that wants toprivatize through cash sales. Gibbon dis-cusses the steps such a government musttake in developing a divestment program.These include setting up a structure forprivatization (including legislation, if nec-essary), providing adequate performancerecords for SOEs being sold (generatingbelievable accounting data), developingany necessary new regulatory structures,and determining the appropriate post-sale relationship between the firm and

the government. Others who examinenon-pricing issues relating to actualdivestment contracts include CarlissBaldwin and Sugato Bhattacharya(1991), Rondinelli and Iacono (1996),Klaus Schmidt (1996), Shafik (1996),and Francesca Cornelli and David Li(1997).

Two empirical papers analyze thechoice of privatization method. One ex-plicitly studies the choice between an as-set sale and a share issue privatization.Using a sample of 1,992 privatizationsthat raised $720 billion in 92 countries,Megginson, Nash, Netter, and AnnettePoulsen (2000) examine why 767 firmswere divested using share offerings (inpublic capital markets), but 1225 com-panies were privatized via direct sales(in private markets). They find robustresults that the choice is influenced bycapital market, political, and firm-specificfactors, and report that SIPs are morelikely to be used when capital marketsare less developed, presumably as a wayto develop capital markets, and whenthere is less income inequality. SIPs arealso more likely the larger the size ofthe offering and the more profitable theSOE. On the other hand, governmentswith greater ability to commit to prop-erty rights are more likely to privatizevia asset sales. Perhaps the most inter-esting result is that governments chooseto privatize the more profitable SOEsthrough SIPs—evidence supporting thepossibility of sample selection bias instudies of performance of privatizedfirms. In the second paper, BernardoBortolotti, Marcella Fantini, andDomenico Siniscalco (1999a) estimatethe determinants of the fraction of pri-vatization revenues that come frompublic offerings (SIPs) for privatizationsin 49 countries. They find that thegreater the selling government’s deficitand the more conservative the sellinggovernment, the more likely it is that

340 Journal of Economic Literature, Vol. XXXIX (June 2001)

privatization will occur through publicofferings. However, SIPs are less likelyin French civil law countries. Bortolotti,Fantini, and Carlo Scapa (2000) exam-ine factors that lead countries to sellshares in SOEs abroad.

4.3 Restructuring SOEs, andSequencing and Staging of Sales

Some of the most complex issues in-volve the interrelated questions ofwhen to privatize and at what pace,what order to follow in privatizing(sequencing), whether to sell an SOEall at once or in stages (staging),whether to restructure an SOE priorto sale, and the role of macroeconomicreform in privatization. Since these arecomplex issues that involve factorsoutside the scope of this article (espe-cially macroeconomic reform which wedo not discuss) we do not spend muchtime on them. Further, their complex-ity has limited empirical work in thisarea.

Several authors have theoreticallymodeled the sequencing and staging ofSOE sales, including Barbara Katz andJoel Owen (1993, 1995), Boycko,Shleifer, and Vishny (1996b), FrancescaCornelli and David Li (1997), EnricoPerotti (1995), and Bruno Biais andPerotti (2000). The models illustratethe importance of sequencing and stag-ing to build reputational capital withinvestors by the privatizing govern-ment, building domestic support forthe program, and identifying biddersthat will maximize the efficiency of thefirm. While the complexities of theseinterrelationships have limited empiri-cists’ ability to identify factors in se-quencing and staging, several articlesthat empirically examine them arePerotti and Serhat Guney (1993),Dewenter and Malatesta (1997), Jones,Megginson, Nash, and Netter (1999),

and Megginson, Nash, Netter, andPoulsen (2000).

A related practical question aboutprivatization is whether governmentsshould restructure SOEs (e.g., lay offredundant workers) prior to selling orleave this to the new owners. This is re-lated to questions discussed in section3.4: can governments reform SOEs (in-cluding reform without privatization)and should reform and privatizationproceed quickly or slowly? Early advicefrom the World Bank (John Nellis andSunita Kikeri 1989) was that govern-ments should restructure SOEs prior todivestment, since governments are bet-ter able than private owners to cushionthe financial blow to displaced workersby using unemployment payments orpensions. Government-led restructuringcan thus provide a private buyer of theSOE with a “clean slate.” Preparingcompanies for privatization was stan-dard practice in the United Kingdomduring the 1980s, in part to smooth thetransition with the trade unions. How-ever, by 1992, the same authors (Kikeri,Nellis, and Shirley 1992) had becomemore nuanced in their interpretation ofthe optimal strategy. They said (p. 54)that small and medium-sized SOEs“should be sold ‘as is’ at the best pricepossible, as quickly as possible.” Theyalso noted that in all cases (p. 60) newinvestments “should be left to privateowners once a decision has been madeto privatise the enterprise.”

Two empirical papers that examineSOE reform prior to privatization areLópez-de-Silanes (1997) and Dewenterand Malatesta (2000). López-de-Silanesexamines whether prior government re-structuring of SOEs improves the netprice received for the company, andfinds evidence that it does not. Heshows that prices would have increasedby 71 cents per dollar of assets if theonly restructuring step taken had been

Megginson and Netter: From State to Market 341

to fire the CEO and if the assets hadbeen divested an average of one yearearlier. He argues that other restructur-ing steps slow down the process andconsume too many resources to beworthwhile. The 71 cents per dollar inadded value would be a significant im-provement on the average 54 cents perdollar of assets actually received. How-ever, this evidence is based on a smallsample of banks, which limits its useful-ness. Dewenter and Malatesta (2001)find some evidence that the improve-ments brought about by privatizationoccur before the SOE is privatized.

4.4 Pricing and Allocation of Controland Ownership

Although mass or voucher privatiza-tion programs have attracted a greatdeal of academic interest, asset salesand SIPs account for most of the valueof assets that have been divested bygovernments in the past two decades.18

Thus we focus on the latter two methods.

4.4.1 Pricing Decisions in Asset Sales

Four papers study the revenue im-pact of SOE direct sale pricing deci-sions. At a theoretical level, Jeremy Bu-low and Paul Klemperer (1996) askwhether it is more profitable to sell acompany through an auction with no re-serve price or by using an optimallystructured direct negotiation with oneless bidder. They show that under mostconditions, a simple competitive auc-tion with N + 1 bidders will yield moreexpected revenue than a seller could ex-pect to earn by fully exploiting his orher monopoly selling position against Nbidders. López-de-Silanes’ (1997) study

of Mexican privatizations empiricallysupports this theoretical conclusion thatmaximizing the number of bidders in anopen auction is usually the best way tomaximize revenues.19 He finds thatprices received are sensitive to the levelof competition in the auction processbut that the Mexican government fre-quently restricted participation (par-ticularly by foreigners) in spite of thisfact. Nonetheless, the amount of reve-nue generated was the main criteria inselecting the winning bidder for morethan 98 percent of the SOEs sold.

Rondinelli and Iacono (1996) exam-ine auctions in central and easternEurope, where thousands of small busi-nesses have been auctioned off, as wellas in Latin America and Russia, wherelarger SOEs have been sold. Manytypes of auctions have been used, in-cluding English, Dutch, first price, sec-ond price, double, and pro-rata sales.Auctions have been used to sell bothlease rights and ownership rights. Inother cases, governments have soldSOEs directly to groups of private in-vestors or firms, setting prices andterms by negotiation. In some cases, thegroups of investors consist of manage-ment or employees. In other cases, thegovernment has liquidated the SOEand sold physical assets to a group ofinvestors.

Archana Hingorani, Kenneth Lehn,and Anil Makhija (1997) examine an ac-tual voucher privatization program, thefirst round of the Czech Republic’smass privatization in 1991. Because themechanics of how companies are di-vested by this government are actuallymore similar to an asset sale than to anyother method, we discuss their workhere. Hingorani, Lehn, and Makhijatest whether the level of share demand,

18 However, it is also true that a much largernumber of companies were transferred to privateownership through mass privatization programs. Itis also likely that more employees were from firmsthat were transferred in mass schemes than fromfirms that were sold in SIPs. We thank John Nellisfor pointing this out to us.

19 The Mexican program relied almost exclu-sively on direct sales, rather than SIPs, as its prin-cipal divestment technique.

342 Journal of Economic Literature, Vol. XXXIX (June 2001)

as measured by voucher redemptions byCzech citizens, effectively predicts theactual level of stock prices in the second-ary market. The authors confirm thepredictive power of share demand, andalso document that share demand ispositively related to the level of insidershareholdings and the extent of foreignownership in a company being sold.They find that share demand is posi-tively related to the level of past profit-ability, which itself shows that even im-perfect accounting statements conveyuseful information. Additionally, they findthat share demand is inversely relatedto the firm’s market risk, which theymeasure as the post-offering coefficientof variation of stock prices.20

4.4.2 Pricing and Share and ControlAllocation in SIPs

Any government that intends to pri-vatize SOEs using public share offer-ings faces three sets of interrelated de-cisions: (1) how to transfer control, (2)how to price the offer, and (3) how toallocate shares. The control transfer de-cision includes whether to sell the SOEall at once or through a series of partialsales. If the government chooses thelatter course, then it must determinehow large a fraction of the company’sshares to issue in the initial versus sub-sequent offers. The government mustalso decide whether to insert any post-privatization restrictions on corporatecontrol. The pricing decision requiresthat the government determine theamount of underpricing, and whetherthe offer price should be set by a tender

offer, a book-building exercise, or at afixed price. If the latter, the govern-ment must decide whether the offeringprice should be set immediately prior tothe offer or many weeks in advance.The share allocation decision requiresthe government to choose whether tofavor one group of potential investorsover another (i.e., domestic investors,SOE employees, or both, over foreignand institutional investors). It also re-quires deciding whether to use the bestavailable investment banker as lead un-derwriter (regardless of nationality) orto favor a national champion.

Several papers empirically examinethe choices governments make in de-signing SIP programs. Kojo Menyahand Krishna Paudyal (1996) and Men-yah, Paudyal, and Charles Inyangete(1995) investigate how the aims and ob-jectives of privatization influence theprocedures and incentives used in thesale of state-owned shares on the Lon-don Stock Exchange by the U.K. gov-ernment. Jones, Megginson, Nash, andNetter (1999), Qi Huang and RichardLevich (1998), and Dewenter andMalatesta (1997) present comprehen-sive studies of the pricing and share andcontrol allocation decisions made bygovernments disposing of SOEs throughpublic share offering. The results arebroadly similar, so we concentrate onthe paper by Jones et al. (1999) since ithas the largest sample.

Jones et al. (1999), whose results aresummarized in table 2, provide evi-dence on the way political factors im-pact the offer pricing, share allocation,and other terms in SIPs. They analyze alarge sample of 630 SIPs from 59 coun-tries made over the period June 1977 toJuly 1997.21 One result they document

20 Stijn Claessens (1997) examines the relationbetween ownership concentration and equityshare prices from the voucher bidding rounds andthe secondary market prices for the 1491 firmsthat emerged from the mass privatization voucherscheme in the Czech and Slovak Republics. Hefinds that the prices are related to the resultingownership structure, with more concentrated own-ership associated with higher prices.

21 Though Jones et al. rely primarily on Privati-sation International for the data used in thisstudy; one of the authors has also developed fromsecondary sources (primarily the Financial Times,

Megginson and Netter: From State to Market 343

is the sheer size of SIP offers—themean (median) size of initial SIPs is$555.7 million ($104.0 million) and themean size of seasoned issues is $1.069billion (median $311.0 million), muchlarger than typical stock offerings. Theyalso find that SIPs are significantly un-derpriced by government sellers. Themean level of underpricing for initial

SIPs is 34.1 percent (median 12.4 per-cent). Even seasoned SIP offers are un-derpriced by an average of 9.4 percent(median 3.3 percent). We return to thisissue in section 6.

The evidence of Jones et al. on allo-cation of control in SIPs supports a po-litical interpretation of divesting gov-ernments’ motives. Jones et al. find thatnearly all SIPs are essentially secondaryofferings, in which only the governmentsells its shares and no money flows tothe firm itself. Since the divesting gov-ernment sells an average (median) of

TABLE 2PRICING, SHARE ALLOCATION, AND CONTROL ALLOCATION PATTERNS IN SIPS

Sample of 630 share issue privatizations (SIPs) executed by 59 national governments during 1977–97. Measuresare broken down for the 417 initial public offerings of SIP shares and the 213 seasoned SIP offerings.

Initial SIPS Seasoned Offers

Measure Mean Median Number Mean Median Number

Pricing VariablesIssue size (US$ million) 555.7 104.0 417 1,068.9 311.0 172Initial return1 34.1 12.4 242 9.4 3.3 55Percent of offer at fixed price2 85.0 100.0 273 61.0 100.0 77Cost of sales as a percent of issue3 4.4 3.3 178 2.5 2.6 61

Share Allocation VariablesPercent of offer allocated to employees 8.5 7.0 255 4.8 2.6 76Fraction of offers with some allocation to

employees 91.0 255 65.8 76Percent of offer allocated to foreigners 28.4 11.5 348 35.9 32.5 142Percent of offers with some allocation to

foreigners 57.1 348 67.6 142

Control Allocation VariablesPercent of capital sold in offer4 43.9 35.0 384 22.7 18.1 154Percent of offers where 100% of capital sold 11.5 384 0 154Percent of capital where 50% or more of

capital sold28.9 384 8.4 154

Source: Jones, Megginson, Nash, and Netter (1999).Notes:1 also known as initial underpricing, the return an investor who bought shares at the offering price could earn byreselling those shares at end of the first day’s trading.2 measures the fraction of an issue offered to investors at a predetermined, fixed price rather than an auction-determined price.3 a measure of the sum of cash expenses and underwriter discount charged by the investment banking syndicatemanaging the issue.4 measures the fraction of a firm’s total common equity (which is not necessarily synonymous with total voting rights)sold in an offering.

but also publications such as Price Waterhouse1989b) an appendix that details similar infor-mation for an additional 500 SIPs. This appen-dix can be obtained upon request by contactingwmegginson@ou.edu.

344 Journal of Economic Literature, Vol. XXXIX (June 2001)

43.9 percent (35.0 percent) of theSOE’s capital in initial offers and 22.7percent (18.1 percent) in seasoned is-sues, the offers represent significant re-ductions in direct government stockownership. Although governments typi-cally surrender day-to-day operatingcontrol of the SOE to private owners inthe initial SIP, they retain effective vetopower through a variety of techniques.The most common technique is govern-ment retention of a “golden share,”which gives it the power to veto certainactions, such as foreign takeovers.22

4.5 Voucher Privatizations

Voucher privatization has been themost controversial method of divestingstate-owned assets. Boycko, Shleifer,and Vishny (1994) show that the decisionto pursue mass privatization, and eventhe specific program design, is largelydictated by politics. The privatizationprograms practiced in western Europeand elsewhere were politically difficultto execute in eastern Europe, althoughHungary, Estonia, and Poland usedcase-by-case privatizations, which havebeen successful at a macro level. None-theless, voucher privatization schemescan be made attractive from an economicperspective, since they maximize value,foster free and efficient markets, andpromote effective corporate governance.

Barbara Katz and Joel Owen (1997)

investigate what they call the “voucherportfolio problem.” This problem ariseswhenever the proportion of ownershipresulting from a given voucher bid isunknown, but the post-privatizationperformance of a divested companylargely depends on the skills of the newowners and their respective ownershipstakes. Katz and Owen also provide agood discussion of the philosophical dif-ferences between the Czech program,which relied heavily on vouchers andprohibited post-sale trading of stock,and the Russian program, which priva-tized relatively small (29 percent on av-erage) stakes in most firms and allowedunrestricted trading of vouchers.

Although most countries’ actual expe-rience with vouchers has been poor,none has been quite as dismal as Rus-sia’s. Although a variety of factors haveplayed a role, Frydman, Katharina Pis-tor, and Rapaczynski (1996) show thatinsider control of privatized firms hasbeen by far the most important impedi-ment to effective reform. Initially, theRussian government had high hopesthat the “voucher privatization funds”(VPFs) formed during the initialvoucher distributions might be able toovercome the collective action probleminherent in mass privatization pro-grams. Such funds might use their con-centrated ownership in privatized firmsto force managers to restructure.Though most funds attempted to exer-cise their “voice” in corporate board-rooms, insider dominance completelyblocked their efforts. The VPFs turnedinstead to their “exit” option and soldshares on the secondary market.

Pistor and Andrew Spicer (1996) alsoexamine the early promise and sub-sequent failure of privatization invest-ment funds in Russia and the Czech Re-public. In both countries, citizens havebecome owners of the worst perform-ing privatized assets, while the “crown

22 Though golden shares have been widelyadopted, they are in fact almost never used to af-fect control contests (Patrick McCurry 2000). TheEU is trying to block new adoptions of goldenshares and roll back those already in place, charg-ing they are designed to discourage free cross-border competition for corporate control. At a re-cent OECD conference, the director of Italy’s pri-vatization program, Vittorio Grilli, pointed out anadditional political problem with exercising agolden share: When a government uses its share toveto a takeover bid, this is equivalent to publiclystating it does not approve of the bidder. Such astatement is awkward at best, and could cause aninternational incident if the bidder is a foreigncompany.

Megginson and Netter: From State to Market 345

jewels” have all come under insidercontrol. As the authors say, “. . . estab-lishing property rights is a longer andmore complicated process than allocat-ing title.” Olivier Blanchard and PhillipeAghion (1996) also conclude that priva-tization is proceeding slowly in easternEurope, largely because insiders, whocurrently have control of firms but noproperty rights, oppose outsider privati-zation. Given this reality, Blanchard andAghion examine whether privatizationwould proceed more rapidly if govern-ments were simply to allocate propertyrights to insiders (insider privatization).However, they find there is a wedge be-tween the private value of the firm toinsiders and its value to an outsider,and that this difference might well pre-clude value-increasing exchanges. Giventhe actual experience with insiderdominance of most voucher privatiza-tions, we conclude that this wedge isin fact alive, well, and fully operational.

5. Has Privatization ImprovedPerformance?

Since privatization has been part ofgovernment policy tool kits for almosttwo decades now, academic researchershave had enough time to execute manyempirical studies of the effect of divest-ment on the performance of formerSOEs. However, there are difficultmethodological problems with researchin this area.23 An important problem isdata availability and consistency. Theamount of information that must be dis-closed is much less in most countriesthan in the United States, and these

standards vary from country to countryas well as within countries over time. Alarge literature in accounting has shownthat management can manipulate U.S.accounting data, and this problem isprobably greater for international firms.Furthermore, the possibility of sampleselection bias can arise from severalsources, including governments’ desireto make privatization “look good” byprivatizing the healthiest firms first. An-other sample selection problem is thatdata availability tends to be greater inthe more developed countries (and per-haps for the better performing firmswithin countries), so developed coun-tries (and better performing firms) areoverrepresented in empirical analysis.For example, in cross-sectional regres-sions using fixed effects, estimation willprobably rely mainly on data fromdeveloped countries.

There are also many problems inmeasuring performance changes thatarise from using accounting or stockdata. We discuss the problems withstock return data in section 6; the prob-lems with accounting data are more im-portant since many empirical studiesemploy primarily accounting informa-tion. These problems include determin-ing the correct measure of operatingperformance, selecting an appropriatebenchmark with which to compare per-formance, and determining the appro-priate statistical tests to use (AhmedGalal, Leroy Jones, Pankaj Tandon, andIngo Vogelsang 1994; and Brad Barberand John Lyon 1996). The finance lit-erature has not reached a consensus onthe ways to deal with these problemsfor U.S. companies, much less priva-tized international firms. Barber andLyon (1996) argue that test statisticsdesigned to determine whether there isabnormal performance using accountingdata are misspecified when the samplefirms have performed unusually well or

23 Many of the difficulties are similar to thosediscussed in Jonathan Temple (1999), who surveyscross-country research on the determinants ofgrowth. Temple discusses the substantial problemsthat arise in estimating and interpreting cross-country regressions. James Tybout (2000) also dis-cusses the difficulties with data in attempting toassess the performance of manufacturing firms indeveloping countries.

346 Journal of Economic Literature, Vol. XXXIX (June 2001)

poorly. They suggest that sample firmsmust be matched to control firms withsimilar pre-event performance, which isespecially difficult in studies of privatizedfirms.

Therefore, the results of each of thestudies we discuss must be kept inperspective. We also note that the stud-ies of post-performance rarely examinethe welfare effects on consumers. Mostimportant, few studies control for thepossible use of market power by theprivatized firms; that is, performanceimprovements could be due to greaterexploitation of monopoly power, whichhas harmful effects on allocative effi-ciency, rather than productive effi-ciency. Many of the studies on perfor-mance changes after privatizationexamine the effects of divestiture ongroups such as workers, but few exam-ine the effect of privatization on con-sumers. On the other hand, one of theprincipal reasons for launching privati-zations, particularly of monopoly utili-ties, is consumer dissatisfaction with afirm’s service. Furthermore, the studiescited here almost unanimously reportincreases in performance associatedwith privatization.24 This consistency isperhaps the most telling result we re-port—privatization appears to improveperformance measured in many differ-ent ways, in many different countries.25

With the above caveats in mind, thissection evaluates the results of 38 stud-

ies that employ accounting and/or realoutput data to examine the impact ofprivatization on the operating effi-ciency, ownership structure, and/or fi-nancial performance of former SOEs indeveloped, developing, and transitioneconomies. Though all these studies aredetailed in the accompanying tables,and most are discussed at least brieflyin the text, we also specify which stud-ies we think are the most important—and why we think this is so. To effec-tively synthesize such a large number ofempirical studies, we first categorizepapers according to whether they exam-ine privatization in transition or non-transition economies. The latter studiesare evaluated in section 5.1, while thetransition economies are examined insection 5.2. This dichotomization is nec-essary, since both direct observationand published research suggest that re-forming transition economies invariablyrequires embracing a great many eco-nomic and political changes simultane-ously, whereas privatization (and atten-dant regulatory changes) is often thesole major component of reform pro-cesses in non-transition economies. Afurther organizational step is to present,in tables 3 through 7, summary informa-tion for each of the studies we examine.Presenting this information in tabularform saves us from having to sequen-tially discuss each paper’s sample con-struction methodology, estimation pro-cedure, and empirical results in thesection’s text. Instead, we can identify keyfindings that appear in many differentstudies, and can discuss methodological prosand cons for entire groups of studies,rather than for each paper in turn.

5.1 Empirical Studies of Non-TransitionEconomies

We separate non-transition studies byempirical methodology, depending uponhow the papers compare performance

24 A cynic might say that all of the gains re-searchers have documented after privatization aredue to selection bias. However, while there issome evidence discussed elsewhere that the betterfirms are privatized first, at least in SIPs the evi-dence is still strong that performance improves af-ter privatization. Further, the paper that does thebest job of controlling for selection bias, Frydman,Gray, Hessel, and Rapaczynski (1999), finds pri-vatized firms perform better than SOEs.

25 Temple (1999) also notes the importance ofboth historical case studies and cross-sectionalanalysis in assessing recent developments in theeconomic theory of growth.

Megginson and Netter: From State to Market 347

changes resulting from privatization.The first set of papers examines a singleindustry, a single country, or one or asmall number of individual firms. Whilethese studies employ a variety of em-pirical techniques, most compare post-privatization performance changes witheither a comparison group of non-privatized firms or with a “counterfac-tual” expectation of what would haveoccurred if the privatized firms had re-mained state-owned. The second set ofstudies examines only firms divestedthrough public share offerings, andmeasures privatization-related perfor-mance changes by comparing the three-year mean or median operating and fi-nancial performance of divested firmsto their own mean or median perfor-mance during their last three years asstate-owned firms.

5.1.1 Case, Single-Industry, andSingle-Country Studies

The studies we examine in this sec-tion are summarized in table 3. Thefirst study listed merits detailed analysisbecause it has proven so influential,both due to the rigor of its methodologyand because it was sponsored by theWorld Bank. Galal, Jones, Tandon, andVogelsang (1994) compare the actualpost-privatization performance of twelvelarge firms—mostly airlines and regu-lated utilities—in Britain, Chile, Malay-sia, and Mexico to the predicted perfor-mance of these firms had they not beendivested. Using this counterfactual ap-proach, the authors document net wel-fare gains in eleven of the twelve casesconsidered which equal, on average, 26percent of the firms’ pre-divestituresales. They find no case where workersare made significantly worse off, andthree where workers significantly bene-fit. David Newberry and Michael Pollitt(1997) perform a similar counterfactualanalysis of the 1990 restructuring and

privatization of the U.K.’s Central Elec-tricity Generating Board (CEGB), anddocument significant post-privatizationperformance improvements. However,they find that the producers and theirshareholders capture all of the financialrewards of this improvement and more,whereas the government and consumerslose out. The authors conclude thatCEGB’s restructuring and privatizationwas in fact “worth it,” but could havebeen implemented more efficiently andwith greater concern for the public’swelfare.26

Two of the studies described in table3 examine national privatization experi-ences. Stephen Martin and DavidParker (1995) find that, after adjustingfor business cycle effects, less than halfthe British firms they study performbetter after being privatized. Theauthors do, however, find evidence of a“shake-out” effect, where several firmsimprove performance prior to being pri-vatized (but not afterward). The resultsof the second national study are far lessambiguous. La Porta and López-de-Silanes (1999) find that the formerMexican SOEs they study rapidly closea large performance gap with industry-matched private firms that had existedprior to divestment. These firms gofrom being highly unprofitable beforeprivatization to being very profitablethereafter. Output increases 54.3 per-cent, in spite of a reduced level of in-vestment spending, and sales per em-ployee roughly double. The privatizedfirms reduce (blue- and white-collar)

26 The privatization and liberalization of theBritish electricity industry is also discussed atlength in David Newberry (1997) and Vickers andYarrow (1991), while the regulatory regimeadopted for earlier utility privatizations is de-scribed in M. E. Beesley and S. C. Littlechild(1989). None of these works showers the Thatchergovernment with praise for its policy decisions,though Beesley and Littlechild do find the RPI-Xprice regulation system adopted in the U.K. issuperior to the U.S. rate of return regulatory regime.

348 Journal of Economic Literature, Vol. XXXIX (June 2001)

TABLE 3CASE STUDIES, COUNTRY AND INDUSTRY-SPECIFIC EMPIRICAL STUDIES: NON-TRANSITION ECONOMIES

Study Sample description, study period, and methodology Summary of findings and conclusions

Galal, Jones,Tandon, andVogelsang1994

Compares actual post-privatization performance of 12large firms (mostly airlines and regulated utilities) inUK, Chile, Malaysia, Mexico to predicted perfor-mance if the firms remained SOEs.

Documents net welfare gains in 11 of the 12 caseswhich equal, on average, 26% of the firms’ pre-divestiture sales. Find no case where workers weremade worse off, and 3 where workers were madesignificantly better off.

Martin andParker 1995

Using 2 measures (ROR on capital employed and an-nual growth in value-added per employee-hour), exam-ines whether 11 UK firms privatized in 1981–88 im-proved performance after divestment. Attempts tocontrol for business cycle effects.

Mixed results. Outright performance improvementsafter privatization found in less than half of firm-measures studied. Several firms improved prior todivestiture, indicating an initial “shake-out” effectupon privatization announcement.

Ramamurti1996

Surveys studies of 4 telecom, 2 airline, and 1 toll-roadprivatization programs in Latin America during 1987–91. Discusses political economic issues, methods usedto overcome bureaucratic/ideological opposition todivestiture.

Concludes privatization very positive for telecoms,partly due to scope for technology, capital investment,and attractiveness of offer terms. Much less scope forproductivity improvements for airlines and roads, andlittle improvement observed.

Boles deBoer andEvans 1996

Estimates impact of 1987 deregulation and 1990privatization of Telecom New Zealand on price andquality of telephone services. Examines whether in-vestors benefited.

Documents significant declines in price of phone ser-vices, due mostly to productivity growth that cut costsat a 5.6% annual rate, and significant improvementin service levels. Shareholders also benefited signifi-cantly.

Petrazziniand Clark1996

Using International Telecommunications Union (ITU)data through 1994, tests whether deregulation andprivatization impact level and growth in teledensity (mainlines per 100 people), prices, service quality, andemployment by telecoms in 26 developing countries.

Deregulation and privatization both are associated withsignificant improvements in level and growth in tele-density, but have no consistent impact on service quality.Deregulation associated with lower prices and increasedemployment; privatization has the opposite effect.

Ramamurti1997

Examines restructuring and privatization ofFerrocarilla Argentinos, the national railroad, in 1990.Tests whether productivity, employment, and needfor operating subsidies (equal to 1% of GDP in 1990)change significantly after divestiture.

Documents a 370% improvement in labor productivityand a 78.7% decline in employment (from 92,000 to19,682). Services were expanded and improved, anddelivered at lower cost to consumers. Need for op-erating subsidies largely eliminated.

Eckel,Eckel, andSingal 1997

Examines effect of British Airways’ privatization oncompetitors’ stock prices. Tests whether fares oncompetitive routes decline after privatization.

Stock prices of US competitors decline on average 7%upon BA’s privatization, and fares on routes servedby BA and competitors fall by 14.3%. Compensationof BA executives increases and becomes moreperformance-contingent.

Newberryand Pollitt1997

Performs cost-benefit analysis of the 1990 restruc-turing and privatization of Central Electricity Generat-ing Board (CEGB). Compares actual performance ofprivatized firms to a counterfactual assuming CEGBremained state-owned.

Restructuring/privatization of CEGB resulted in per-manent cost reduction of 5% per year. Producers andshareholders capture all this benefit and more. Con-sumers and government lose. Shows that alternativefuel purchases involve unnecessarily high costs andwealth flows out of country.

Ros 1999 Uses ITU data and panel data regression methodologyto examine effects of privatization and competition onnetwork expansion and efficiency in 110 countriesover 1986–95.

Countries with at least 50% private ownership of maintelecom firm have significantly higher teledensitylevels and growth rates. Both privatization and com-petition increase efficiency, but only privatization ispositively associated with network expansion.

La Porta andLópez-de-Silanes 1999

Tests whether performance of 218 Mexican SOEsprivatized through June 1992 improves after di-vestment. Compares performance with industry-matched firms, and splits improvements documentedbetween industry and firm-specific influences.

Output of privatized firms increased 54.3%; employ-ment declined by half (though wages for remainingworkers increased). Firms achieved a 24% point in-crease in operating profitability, eliminating need forsubsidies equal to 12.7% of GDP. Higher productprices explain 5% of improvement; transfers from laid-off workers, 31%, and incentive-related productivitygains account for remaining 64%.

Megginson and Netter: From State to Market 349

employment by half, but those workerswho remain are paid significantly more.The authors attribute most of the per-formance improvement to productivity gainsresulting from better incentives, with atmost one-third of the improvement beingattributable to lower employment costs.

Three of the papers described in ta-ble 3 are essentially case studies of indi-vidual privatized companies, thoughtwo of the articles benchmark perfor-mance changes with respect to one ormore private companies. Catherine Eckel,Doug Eckel, and Vijay Singal (1997) ex-amine the effect of British Airways’ (BA)1987 privatization on competitors’ stockprices and on fares charged in thoseroutes where BA competes directly withforeign airlines. They find that thestock prices of U.S. competitors fall, asdo airfares in markets served by BA;both findings suggest that stock tradersanticipated a much more competitive

BA would result from the divestiture.27

Claude Laurin and Yves Bozec (2000)compare the productivity and profit-ability of two large Canadian rail carri-ers (one state-owned and one private-sector), both before and after the 1995privatization of Canadian National(CN). They find that CN’s relativelypoor performance during the “fullystate-owned period” (1981–91) rapidlyconverges on Canadian Pacific’s perfor-mance levels during the pre-privatizationbut post-announcement period (1992–95),and then surpasses it thereafter. These

TABLE 3 (Cont.)

Study Sample description, study period, and methodology Summary of findings and conclusions

Wallsten2000a

Performs econometric analysis of effects of tele-communications reforms in developing countries.Using panel dataset of 30 African and Latin Americancountries from 1984–97, explores effects of pri-vatization, competition and regulation on telecom-munications performance.

Competition is significantly associated with increasesin per capita access and decreases in cost. Privatizationis helpful only if coupled with effective, independentregulation. Increasing competition is single best re-form; competition in combination with privatization isbest. Privatizing a monopoly without regulatory re-forms should be avoided.

Laurin andBozec 2000

Compares productivity and profitability of 2 largeCanadian rail carriers, before and after 1995 pri-vatization of Canadian National (CN). Compares ac-counting ratios for 17-year period 1981–97 and 3sub-periods: the fully state-owned era (1981–91), pre-privatization (1992–95), and post-privatization. Com-pares stock returns from 1995–98. Creates 6-firmcomparison group of Canadian privatizations and com-putes accounting ratios and stock returns for thesefirms.

Total factor productivity of CN much lower than thatof privately owned Canadian Pacific (CP) during 1981–91, but became as efficient during pre-privatization(1992–95), exceeded it after 1995. CN stock priceoutperformed CP, the transportation industry, and theCanadian market after 1995. Both firms shed workersafter 1992, but CN’s employment declined more (34%vs 18%) as average productivity almost doubled (97%increase). CN’s capital spending increased signifi-cantly, though CP’s increased more. Six-firm Canadianprivatization comparison group experienced significantincreases in investment spending and productivity anddecline in employment.

Boylaud andNicoletti2000

Uses factor analysis and database on market structureand regulation to investigate effects of liberalizationand privatization on productivity, prices and quality oflong-distance and cellular telephone services in 23OECD countries over 1991–97.

Prospective and actual competition both bring aboutproductivity and quality improvements and lowerprices in telecom services, but no clear effect wasfound for privatization.

27 Eckel, Eckel, and Singal also examine thetwo-stage privatization of Air Canada (from 100percent state ownership to 57 percent, then tozero). Unlike BA, Air Canada does not competewith U.S. carriers on many routes, so there is nosignificant competitor stock price effect resultingfrom its divestiture. Air Canada’s fares do not fallafter the first, partial privatization, but fall a sig-nificant 13.7 percent after the final, complete di-vestiture of state ownership.

350 Journal of Economic Literature, Vol. XXXIX (June 2001)

findings suggest two separable impactsof privatization on firm performance: an“anticipation” effect prior to divestitureand a “follow through” effect sub-sequently. The final case study, RaviRamamurti (1997), examines the 1990restructuring and privatization of Ferro-carilla Argentino, the Argentine na-tional freight and passenger railway sys-tem. The author documents a nearlyincredible 370 percent improvement inlabor productivity and an equally strik-ing (and not unrelated) 78.7 percent de-cline in employment—from 92,000 to18,682 workers.28 Operating subsidiesdeclined almost to zero, and consumersbenefited from expanded (and betterquality) service and lower costs. Rama-murti concludes that these performanceimprovements could not have beenachieved without privatization.

No less than six of the studies de-tailed in table 3 examine the telecom-munications industry, which has beentransformed by the twin forces of tech-nological change and deregulation (in-cluding privatization) since 1984—theyear when the AT&T monopoly wasbroken up in the United States and theThatcher government began privatizingBritish Telecom. Five of these are em-pirical studies, while Ramamurti (1996)provides a simple, though highly read-able, summary of empirical studies ex-amining four telecom privatizations inLatin America. Ramamurti concludesthat all were judged to be political andeconomic success stories. Unfortu-nately, the empirical studies tell some-what conflicting stories, probably due inpart to differences in the nations cov-

ered and methodology employed. BenPetrazini and Theodore Clark (1996),Agustin Ros (1999), and Scott Wallsten(2000a) examine developing countriesexclusively or as separate subsamples,while Ros (1999) and Olivier Boylaudand Giuseppe Nicoletti (2000) providesimilar coverage of OECD countries,and David Boles de Boer and LewisEvans (1996) study the deregulationand privatization of Telecom New Zea-land. Though Ros, Wallsten, and Boy-laud and Nicoletti all use some variantof panel data methodology, they arrive atslightly different conclusions regardingthe relative importance of deregulation/liberalization and privatization in pro-moting expanded teledensity (numberof main lines per 100 population) andoperating efficiency of national telecomcompanies, and the quality and pricingof telecom services. On balance, thesestudies generally indicate that deregula-tion and liberalization of telecom ser-vices are associated with significantgrowth in teledensity and operating ef-ficiency, and significant improvementsin the quality and price of telecomservices. The impact of privatization,per se, is somewhat less clear-cut, butmost studies agree that the combina-tion of privatization and deregulation/liberalization is associated with signifi-cant telecommunications improvements.This is certainly the result predicted byNoll (2000) in his analysis of the politi-cal economy of telecom reform in de-veloping countries. The Juliet D’Souzaand Megginson (2000) study’s findings—described in the following section—alsosupport the idea that telecom privatizationyields net benefits.29

28 Ramamurti (1997) details the intense politicalmaneuvering that accompanied the attempt to re-structure and slim down FA. The generous sever-ance payments awarded to displaced workers wereinstrumental in winning union acquiescence in therestructuring plan, while the presence of effectiveroad transport competition for rail traffic reducedthe threat of a potentially crippling strike weapon.

29 Though they do not quite fit into our empiri-cal classification scheme, six related studies de-serve mention here. Peter Smith and Björn Wel-lenius (1999) and Wellenius (2000) presentnormative analyses of telecom regulation in devel-oping countries, while Walter Wasserfallen andStefan Müller (1998) discuss the privatization and

Megginson and Netter: From State to Market 351

5.1.2 Pre- vs Post-PrivatizationPerformance for SIPs

The studies summarized in table 4 allexamine how privatization affects firmperformance by comparing pre- andpost-divestment data for companies pri-vatized via public share offering. Sincethe first study to be published usingthis methodology is Megginson, Nash,and van Randenborgh (1994), we willrefer to this as the MNR methodology.This empirical procedure has severalobvious economic and econometricdrawbacks. Of these, selection biasprobably causes the greatest concern,since by definition a sample of SIPs willbe biased towards the very largest com-panies sold during any nation’s privati-zation program. Furthermore, sincegovernments have a natural tendency toprivatize the “easiest” firms first, thoseSOEs sold via share offerings (particu-larly those sold early in the process)may well be among the healthiest state-owned firms.30 Another drawback of the

MNR methodology is its need to exam-ine only simple, universally available ac-counting variables (such as assets, sales,and net income) or physical units suchas number of employees. Obviously,researchers must be careful whencomparing accounting informationgenerated at different times in manydifferent countries. Most of the stud-ies cited here also ignore (or, at best,imperfectly account for) changes inthe macroeconomy or industry overthe seven-year event window duringwhich they compute pre- versus post-privatization performance changes. Fi-nally, the studies cannot account for theimpact on privatized firms of any regu-latory or market-opening initiatives thatoften are launched simultaneously withor immediately after major privatizationprograms.

In spite of these drawbacks, studiesemploying the MNR methodology havetwo key advantages. First, they are theonly studies that can examine and di-rectly compare large samples of eco-nomically significant firms, from differ-ent industries, privatized in differentcountries, over different time periods.Since each firm is compared to itself (afew years earlier) using simple, inflation-adjusted sales and income data (thatproduce results in simple percentages),this methodology allows one to effi-ciently aggregate multinational, multi-industry results. This point is madeclear in table 5, which summarizes theresults of three studies that use pre-cisely the same empirical proxies andtest methodology—and can thus be ag-gregated and directly compared—yetexamine non-overlapping samples. Intotal, these three studies examine sevenperformance criteria for 204 companiesfrom 41 countries. Second, while focus-ing on SIPs yields a selection bias, italso yields samples that encompass thelargest and most politically influential

deregulation of western Europe’s telecom indus-try. Michael Pollitt (1997) analyzes the impact ofliberalization on the performance of the interna-tional electric supply industry, and Bortolotti,Fantini, and Siniscalso (1999b) document that ef-fective regulation is a crucial institutional variablein electric utility privatization. Establishing such aregulatory regime allows governments to increasethe pace of privatization, sell higher stakes, andmaximize offering proceeds. Finally, Wallsten(2000b) shows that exclusivity periods, which areusually granted to telecom monopolies as theyare being privatized, are economically harmful toconsumers and do not achieve the efficiency ob-jectives assigned to them at the time of divest-ment. Exclusivity periods do, however, raise theprice that investors are willing to pay for priva-tized telecoms, which largely explains why theyare employed.

30 Megginson, Nash, Netter, and Poulsen (2000)find that governments selling SOEs tend to sellthe more profitable SOEs in the public capitalmarkets and the less profitable in the less trans-parent private markets. Those sold in the publiccapital markets are the firms that appear in studiesof performance. Dewenter and Malatesta (2001)also show performance improvements before pri-vatization in firms that are being privatized.

352 Journal of Economic Literature, Vol. XXXIX (June 2001)

TABLE 4EMPIRICAL STUDIES ON PERFORMANCE CHANGES FOR FIRMS PRIVATIZED VIA PUBLIC SHARE OFFERINGS:

NON-TRANSITION ECONOMIES

These studies each employ samples from more than one country and more than one industry.

Study Sample description, study period, and methodology Summary of findings and conclusions

Megginson,Nash, vanRandenborgh1994

Compares 3-year average post-privatization per-ormance ratios to 3-year pre-privatization values for 61firms from 18 countries and 32 industries from 1961–89. Tests significance of median changes in post versuspre-privatization periods. Binomial tests for percent offirms changing as predicted.

Documents economically and statistically significantpost-privatization increases in output (real sales),operating efficiency, profitability, capital investmentspending, and dividend payments; significant de-creases in leverage; no evidence of employment de-clines, but significant changes in firm directors.

Macquieiraand Zurita1996

Compares pre- versus post-privatization performanceof 22 Chilean firms privatized over 1984–89. UsesMegginson, Nash and van Randenborgh (MNR)methodology to analyze first without adjusting foroverall market movements (as in MNR), then withadjustment for contemporaneous changes.

Unadjusted results virtually identical to MNR: sig-nificant increases in output, profitability, employment,investment, dividend payments. After adjusting formarket movements, changes in output, employment,and liquidity are no longer significant, and leverageincreases significantly.

Boubakriand Cosset1998

Compares 3-year average post-privatization per-formance ratios to 3-year pre-privatization values for79 firms from 21 developing countries and 32 indus-tries over 1980–92. Tests for significance of medianchanges in ratio values post- versus pre-privatization.Binomial tests for percentage of firms changing aspredicted.

Documents significant post-privatization increases inoutput (real sales), operating efficiency, profitability,capital investment spending, dividend payments,employment; significant decreases in leverage. Per-formance improvements are generally larger thanthose documented by MNR.

D’Souza andMegginson1999

Documents offering terms, sale methods, andownership structure resulting from privatization of 78firms from 10 developing and 15 developed countriesover 1990–94. Compares 3-year average post-privatization performance ratios to 3-year pre-privatization values for subsample of 26 firms. Testsfor significance of median changes in ratio values post-vs pre-privatization. Binomial tests for percent of firmschanging as predicted.

Documents significant post-privatization increases inoutput (real sales), operating efficiency, and profit-ability, and significant decreases in leverage. Capitalinvestment spending increases insignificantly, whileemployment declines significantly. More of the firmsprivitized in the 1990s are from telecoms and otherregulated industries.

Verbrugge,Megginson,Owens 2000

Study offering terms and share ownership results for65 banks fully or partially privatized from 1981 to1996. Then compare pre- and post-privatization per-formance changes for 32 banks in OECD countriesand 5 in developing countries.

Documents moderate performance improvements inOECD countries. Ratios proxying for profitability, feeincome (noninterest income as fraction of total), andcapital adequacy increase significantly; leverage ratiodeclines significantly. Documents large, ongoing stateownership, and significantly positive initial returns toIPO investors.

Boubakriand Cosset1999

Examine pre- versus post-privatization performance of16 African firms privatized through public share of-fering during 1989–96. Also summarize findings ofthree other studies pertaining to privatization indeveloping countries.

Document significantly increased capital spending byprivatized firms, but find only insignificant changes inprofitability, efficiency, output and leverage.

D’Souza andMegginson2000

Examines pre- versus post-privatization performancechanges for 17 national telecom companies privatizedthrough share offerings during 1981–94.

Profitability, output, operating efficiency, capitalspending, number of access lines, and average salaryper employee all increase significantly after pri-vatization. Leverage declines significantly; employ-ment declines insignificantly.

DewenterandMalatesta2001

Compares pre- versus post-privatization performanceof 63 large, high-information companies divested dur-ing 1981–94 over both short-term [(+1 to +3) versus(−3 to −1)] and long-term [(+1 to +5) versus (−10 to −1)]horizons. Examines long-run stock return performanceof privatized firms and compares relative performanceof a large sample (1,500 firm-years) of state and pri-vately owned firms during 1975, 1985, and 1995.

Documents significant increases in profitability (usingnet income) and significant decreases in leverage andlabor intensity (employees ÷ sales) over short- andlong-term horizons. Operating profits increase prior toprivatization, but not after. Significantly positive long-term (1–5 years) abnormal stock returns, mostly inHungary, Poland, and UK. Results strongly indicatethat private firms outperform SOEs.

Megginson and Netter: From State to Market 353

privatizations. As discussed in section 4,SIPs account for more than two-thirdsof the $1 trillion of total revenuesraised by governments since 1977.With these methodological caveats inmind, we turn to a summary of thefindings of studies using the MNRtechnique.

All of these studies offer at least lim-ited support for the proposition thatprivatization is associated with signifi-cant improvements in the operating andfinancial performance of SOEs divestedvia public share offering. Two of thesestudies focus on specific industries:banking (James Verbrugge, WandaOwens, and Megginson 2000) and tele-communications (D’Souza and Meggin-son 2000); one examines data from asingle country, Chile (Carlos Mac-quieira and Salvador Zurita 1996); andthe other six employ multi-industry,multinational samples. Five of thesestudies—MNR (1994), Narjess Boubakriand Jean-Claude Cosset (1998), D’Souzaand Megginson (1999, 2000), andBoardman, Laurin and Vining (2000)—document economically and statisticallysignificant post-privatization increasesin real sales (output), profitability, effi-ciency (sales per employee), and capitalspending, coupled with significant de-clines in leverage. Macquieira andZurita find similar results for Chileanfirms using data that is not adjusted forchanges experienced by other Chilean

firms over the study period, but manyof these improvements cease to be sta-tistically significant once such adjust-ments are made. Verbrugge et al.(2000) document significant, thoughmodest, increases in the profitabilityand capital adequacy of commercialbanks privatized in OECD countries, aswell as significant declines in leverage,but they also find substantial ongoingstate involvement in these banks’ af-fairs. Consistent with the result thatstate connections matter in bank opera-tions, Philip Hersch, David Kemme,and Netter (1997) find that in Hungarythe banks made it much easier for firmsheaded by former members of the no-menklatura to get loans than otherfirms.

Finally, Dewenter and Malatesta(2000) estimate the effects of govern-ment ownership and privatization usinga sample of large firms from three sepa-rate time periods (1975, 1985, and1995) compiled by Fortune. They esti-mate regressions explaining profitabilitycontrolling for firm size, location, in-dustry, and the business cycle. Theyfind that net income-based profitabilitymeasures increase significantly afterprivatization, but operating income-based measures do not. Instead, theyfind that operating profits increaseprior to divestiture, once more support-ing the idea that privatization can havea significant anticipation effect.

TABLE 4 (Cont.)

Study Sample description, study period, and methodology Summary of findings and conclusions

Boardman,Laurin, andVining 2000

Compares 3-year average post-privatization per-formance ratios to 5-year pre-privatization values for 9Canadian firms privatized during 1988–95. Computeslong-run (up to 5 years) stock returns for divestedfirms.

Profitability, measured as return on sales or assets,more than doubles after privatization; efficiency andsales increase significantly (though less drastically).Leverage and employment decline significantly;capital spending increases significantly. Privatizedfirms significantly outperform Canadian stock marketover all long-term holding periods.

354 Journal of Economic Literature, Vol. XXXIX (June 2001)

TABLE 5PERFORMANCE OF NEWLY PRIVATIZED FIRMS

Results of three empirical studies comparing three-year average operating and financial performance of a combined sample of 211privatized firms with average performance of those firms during their last three years as SOEs. The studies employ the Wilcoxonrank sum test (with its z-statistic) to test for change in median value, and multiple proxies for most economic variables beingmeasured. This table summarizes one proxy per topic, and emphasizes the one highlighted in the studies (usually the variable thatuses either physical measures, such as number of employees, or financial ratios using current-dollar measures in the numerator ordenominator, or both). Efficiency and output measures are index values, with the value during the year of privatization defined as1.000; inflation-adjusted sales figures are used in efficiency and output measures.

Variables andStudies Cited

Number ofObserva-

tions

MeanValuebefore

Privatiza-tion

MeanValue afterPrivatiza-

tion

MeanChangedue to

Privatiza-tion

Z-Statisticfor

Differencein Perfor-

mance

% of Firmswith

ImprovedPerfor-mance

Z-Statisticfor

Significanceof %

Change

Profitability (%) net income ÷ salesMegginson, Nash, and van 55 0.0552 0.0799 0.249 3.15∗∗∗ 69.1 3.06∗∗∗

Randenborgh 1994 (0.0442) (0.0611) (0.0140)Boubakri and Cosset 78 0.0493 0.1098 0.0605 3.16∗∗∗ 62.8 2.29∗∗

1998 (0.0460) (0.0799) (0.0181)D’Souza and Megginson 78 0.14 0.17 0.03 3.92∗∗∗ 71 4.17∗∗∗

1999 (0.05) (0.08) (0.03)Weighted Average 218a 0.0862 0.1257 0.0396 67.6

Efficiency (real sales per employee)Megginson, Nash, and van 51 0.956 1.062 0.1064 3.66∗∗∗ 85.7 6.03∗∗∗

Randenborgh 1994 (0.942) (1.055) (0.1157)Boubakri and Cosset 56 0.9224 1.1703 0.2479 4.79∗∗∗ 80.4 4.60∗∗∗

1998 (0.9056) (1.1265) (0.2414)D’Souza and Megginson 63 1.02 1.23 0.21 4.87∗∗∗ 79 5.76∗∗∗

1999 (0.87) (1.16) (0.29)Weighted Average 170 0.9733 1.1599 0.1914 81.5

Investment (%) capital expenditures ÷ salesMegginson, Nash, and van 43 0.1169 0.1689 0.0521 2.35∗∗ 67.4 2.44∗∗

Randenborgh 1994 (0.0668) (0.1221) (0.0159)Boubakri and Cosset 48 0.1052 0.2375 0.1322 2.28∗∗ 62.5 1.74∗

1998 (0.0649) (0.1043) (0.0137)D’Souza and Megginson 66 0.18 0.17 −0.01 0.80 55 0.81

1999 (0.11) (0.10) (−0.01)Weighted Average 154 0.1405 0.1900 0.0493 60.6

Output (real sales, adjusted by cpi)Megginson, Nash, and van 57 0.899 1.140 0.241 4.77∗∗∗ 75.4 4.46∗∗∗

Randenborgh 1994 (0.890) (1.105) (0.190)Boubakri and Cosset 78 0.9691 1.220 0.2530 5.19∗∗∗ 75.6 4.58∗∗∗

1998 (0.9165) (1.123) (0.1892)D’Souza and Megginson 85 0.93 2.70 1.76 7.30∗∗∗ 88 10.94∗∗∗

1999 (0.76) (1.86) (1.11)Weighted Average 209a 0.9358 1.7211 0.8321 80.3

Employment (total employees)Megginson, Nash, and van 39 40,850 43,200 2,346 0.96 64.1 1.84∗

Randenborgh 1994 (19,360) (23,720) (276)Boubakri and Cosset 57 10,672 10,811 139 1.48 57.9 1.19

1998 (3,388) (3,745) (104)D’Souza and Megginson 66 22,941 22,136 −805 −1.62 36 −2.14∗∗

1999 (9,876) (9,106) (−770)Weighted Average 162 22,936 23,222 286 49.5

Megginson and Netter: From State to Market 355

5.1.3 Summary and Analysis

These 22 studies from non-transitioneconomies offer at least limited supportfor the proposition that privatization isassociated with improvements in theoperating and financial performance ofdivested firms. Several of the studies of-fer strong support for this proposition,and only Martin and Parker (1995)document outright performance de-clines (for six of eleven British firms)after privatization. Almost all studiesthat examine post-privatization changesin output, efficiency, profitability, capitalinvestment spending, and leverage docu-ment significant increases in the first fourand significant declines in leverage.

The studies examined here are farless unanimous regarding the impact ofprivatization on employment levels in

privatized firms. All governments fearthat privatization will cause formerSOEs to shed workers, and the keyquestion in virtually every case iswhether the divested firm’s sales willincrease enough after privatization tooffset the dramatically higher levels ofper-worker productivity. Three studiesdocument significant increases in em-ployment (Galal et al. 1994; Megginson,Nash, and van Randenborgh 1994; andBoubakri and Cosset 1998), two find in-significant changes (Macquieira andZurita 1996; and D’Souza and Meggin-son 2000) while the remaining fivedocument significant—sometimes mas-sive—employment declines (Ramamurti1997; La Porta and López-de-Silanes 1999;Laurin and Bozec 2000; D’Souza and Meg-ginson 1999; and Boardman, Laurin, andVining 2000). These conflicting results

TABLE 5 (Cont.)

Variables andStudies Cited

Number ofObserva-

tions

MeanValuebefore

Privatiza-tion

MeanValue afterPrivatiza-

tion

MeanChangedue to

Privatiza-tion

Z-Statisticfor

Differencein Perfor-

mance

% of Firmswith

ImprovedPerfor-mance

Z-Statisticfor

Significanceof %

Change

Leverage (%)total debt ÷ total assets

Megginson, Nash and van 53 0.6622 0.6379 −0.0243 −2.41∗∗ 71.7 3.51∗∗∗Randenborgh 1994 (0.7039) (0.6618) (−0.0234)

Boubakri and Cosset 65 0.5495 0.4986 −0.508 −2.48∗∗ 73.1 2.11∗∗1998 (0.5575) (0.4789) (−0.0162

D’Souza and Megginson 72 0.29 0.23 −0.06 −3.08∗∗∗ 67 3.05∗∗∗1999 (0.26) (0.18) (−0.08)Weighted Average 188 0.4826 0.4357 −0.0469 67.0

Dividends (%)cash dividends ÷ sales

Megginson, Nash and van 39 0.0128 0.0300 0.0172 4.63∗∗∗ 89.7 8.18∗∗∗Randenborgh 1994 (0.0054) (0.0223) (0.0121)

Boubakri and Cosset 67 0.0284 0.0528 0.0244 4.37∗∗∗ 76.1 4.28∗∗∗1998 (0.0089) (0.0305) (0.0130)

D’Souza and Megginson 51 0.015 0.04 0.025 4.98∗∗∗ 79 5.24∗∗∗1999 (0.00) (0.02) (0.02)Weighted Average 106 0.0202 0.0655 0.0228 80.4

a Number exceeds 211 because of overlapping firms in different samples.∗∗∗ Indicates significance at the 1 percent level

∗∗ Indicates significance at the 5 percent level∗ Indicates significance at the 10 percent level

356 Journal of Economic Literature, Vol. XXXIX (June 2001)

could be due to differences in method-ology, sample size and make-up, oromitted factors. However, it is morelikely that the studies reflect real differ-ences in post-privatization employmentchanges between countries and be-tween industries. In other words, thereis no “standard” outcome. Perhaps thesafest conclusion we can assert is thatprivatization does not automaticallymean employment reductions in divestedfirms—though this will likely occurunless sales can increase fast enoughafter divestiture to offset very largeproductivity gains.

In our opinion, the Galal et al.(1994), La Porta and López-de-Silanes(1999), Dewenter and Malatesta (2001),and the three articles summarized inD’Souza and Megginson (1999) are themost persuasive studies examined inthis section. As mentioned, the mainstrength of Galal et al. is its construc-tion and use of a clear “counterfactual”that (virtually uniquely) allows both thefinancial and welfare gains from privati-zation to be measured. La Porta andLópez-de-Silanes execute what we con-sider the best single-country study,since it examines almost the entirepopulation of Mexican privatizationsand compares performance changes toindustry-matched private firms. Dewen-ter and Malatesta both contrast theperformance of private-sector andstate-owned firms over three non-overlapping periods and study how theperformance of privatized firms changesover an extended time period. Finally,D’Souza and Megginson’s summary andcomparison of three studies that use thesame methodology—but non-overlappingsamples—provides compelling evidencethat the operating and financial gains toprivatization are pervasive.

Since the empirical studies discussedin this section generally document per-formance improvements after privatiza-

tion, a natural follow-on question is toask why performance improves. As wewill discuss in the next section, a keydeterminant of performance improve-ment in transition economies is bring-ing in new managers after privatization.No study explicitly documents system-atic evidence of this occurring in non-transition economies, but CatherineWolfram (1998) and Michael Cragg andI. J. Alexander Dyck (1999a,b) show thatthe compensation and pay-performancesensitivity of managers of privatizedU.K. firms increases significantly after di-vestment. The only study that explicitlyaddresses the sources of post-privatizationperformance improvement using datafrom multiple non-transition econo-mies, D’Souza et al. (2000), findsstronger efficiency gains for firms indeveloping countries, in regulated in-dustries, in firms that restructure op-erations after privatization, and in coun-tries providing greater amounts ofshareholder protection.

We now turn to an examination of re-search findings about privatization’s im-pact in transition economies. Privatiza-tion is both more difficult and moreall-encompassing in these countriesthan it is in either industrialized ornon-transition developing countries.This is because in transition economies,privatization is only part of the massivechanges in the economy as countries movefrom communism to more market ori-ented methods of allocating resourcesand organizing production.

5.2 Privatization in TransitionEconomies

We categorize the empirical studiesthat examine privatization in transitioneconomies into more manageablegroups. Both direct observation and thefindings of these studies suggest that alogical classification scheme is to evalu-ate separately studies that examine

Megginson and Netter: From State to Market 357

firms privatized in central and easternEurope and those which study the pri-vatization programs of Russia and theother republics of the former SovietUnion. These categories are evaluatedin sections 5.2.1 and 5.2.2, respectively.We then conclude section 5 with a briefoverview of China’s liberalization andprivatization program.

Note that testing for the effects ofprivatization on firm performance iseven more difficult in transition econo-mies than in non-transition economies.As mentioned above, privatization inthese countries occurs at the same time as,and is part of, other massive economy-wide changes. Thus, isolating the ef-fects of privatization itself is problem-atic. Further, as discussed by Djankovand Murrell (2000b, p. 9) “mis-reportingand accounting difficulties are rife intransition economies.” In general, thedata from transition economies is muchworse and much more limited thanfrom non-transition economies. Finally,the transition economies are under-going many other major changes in theirpolitical and economic environments.The number of firms privatized in someway in transition is much greater thanin non-transition economies (Djankovand Murrell 2000a report over 150,000large firms in 27 transition economiesfaced the revolutionary changes of tran-sition). However, we do not have gooddata or even any data on many of thesefirms. The data that do exist often comefrom surveys rather than mandated dis-closure. Thus, the studies of privatiza-tion in transition economies has greaterproblems with significant selection bias,as well as omitted variables, than in thestudies of non-transition economies.

5.2.1 Privatization in Central andEastern Europe

The empirical studies that examineprivatization programs in central and

eastern Europe are summarized in table6. These countries employed varyingmethods of privatizing SOEs, includingasset sales (Hungary and eastern Ger-many), voucher privatizations (theCzech Republic and early Polish dives-titures), “spontaneous privatizations”(Slovenia), share offerings (later Polishsales), or a combination of techniques.The studies also cover differing eventperiods during the 1990s, employ dif-fering empirical methodologies, and asksomewhat different questions—thoughall directly or indirectly ask how privati-zation impacts firm-level operating per-formance. Additionally, all of thesestudies must contend with the fact thatoutput typically fell dramatically inevery central and eastern Europeancountry during the period immediatelyafter the collapse of socialism in 1989–91, though in most cases output latersnapped back smartly.31 These studiesmust therefore examine whether, forexample, the output of privatized firmscontracted less than did the output offirms that remained state-owned. Theseand other econometric challenges thatmust be faced in disentangling the ef-fects of privatization, ownership struc-ture changes, and other influences onthe post-divestment performance of pri-vatized firms in transition settings arediscussed at length in Andrew Weissand Georgiy Nikitin (1998) and Frydman,Gray, Hessel, and Rapaczynski—hereafterFGHR—(1999).

In spite of all the caveats spelled outabove, the studies summarized in table6 yield surprisingly consistent results

31 This “U-shape” pattern of aggregate output in26 transition economies is documented and exam-ined econometrically in Andrew Berg, EduardoBorensztein, Ratna Sahay and Jeromin Zettel-meyer (1999). They find that structural reforms—including privatization—are critically important inpromoting rapid recovery from the initial eco-nomic decline. Taken as a whole, their resultsstrongly support a “radical” approach to reforms.

358 Journal of Economic Literature, Vol. XXXIX (June 2001)

TABLE 6SUMMARY OF EMPIRICAL STUDIES OF PRIVATIZATION IN TRANSITION ECONOMIES: CENTRAL AND EASTERN EUROPE

Study Sample description, study period, and methodology Summary of empirical findings and conclusions

Claessens,Djankov, andPohl 1997

Examines determinants of performance improvementsfor 706 Czech firms privatized during 1992–95. UsingTobins-Q, tests whether concentrated ownershipstructure or outside monitor (bank or investmentfund) improves Q more than dispersed ownership.

Privatized firms do prosper, primarily because ofresulting concentrated ownership structure. The moreconcentrated the post-privatization ownershipstructure the higher the firm’s profitability and marketvaluation. Large stakes owned by bank-sponsoredfunds and strategic investors are particularly value-enhancing.

Pohl,Anderson,Claessens,and Djankov1997

Compares extent of restructuring of over 6,300 privateand state-owned firms in 7 east European countriesduring 1992–95. Uses 6 measures to examine whichstrategies improve performance the most.

Privatization dramatically increases restructuringlikelihood and success. Firms privatized for 4 yearswill increase productivity 3–5 times more than similarSOEs. Little difference in performance based onmethod of privatization, but ownership and financingeffects impact restructuring.

Smith, CinandVodopivec1997

Using a sample with 22,735 firm-years of data drawnfrom period of “spontaneous privatization” in Slovenia(1989–92), examines impact of foreign and employeeownership on firms.

Percentage point increase in foreign ownership isassociated with a 3.9% increase in value added, andfor employee ownership with a 1.4% increase. Firmswith higher revenues, profits, and exports are morelikely to exhibit foreign and employee ownership.

Dyck 1997 Develops and tests an adverse selection model to ex-plain Treuhand’s role in restructuring and privatizingeast Germany’s SOEs. In less than 5 years, Treuhandprivatized more than 13,800 firms and parts of firmsand, uniquely, had resources to pay for restructuringitself—but almost never chose to do so. Instead, itemphasized speed and sales to existing western firmsover giveaways and sales to capital funds. Paper ra-tionalizes Treuhand’s approach.

Privatized east German firms were more likely to putwestern (usually German) managers in key positionsthan were companies that remained state-owned.Treuhand emphasized sales open to all buyers ratherthan favoring east Germans. Principal message:privatization programs must carefully consider whenand how to affect managerial replacement in firms.Plans open to western buyers and which allow man-agement change are most likely to improve firm per-formance.

Frydman,Gray, HesselandRapaczynski1999

Compares performance of privatized and state-ownedfirms in central European transition economies, andasks “when does privatization work?” Examines in-fluence of ownership structure on performance usinga sample of 90 state-owned and 128 privatized com-panies in Czech Republic, Hungary, and Poland.Employs panel data regression methods to isolateownership effects.

Privatization “works,” but only when firm is controlledby outside owners (other than managers or em-ployees). Privatization adds over 18 percentage pointsto annual growth rate of firm sold to domestic finan-cial firm, and 12 percentage points when sold to aforeign buyer. Privatization to an outside owner alsoadds about 9 percentage points to productivity growth.Gain does not come at expense of higher unemploy-ment; insider controlled firms are less likely to restruc-ture, but outsider-controlled firms grow faster. Showsthe importance of entrepreneurship in reviving salesgrowth.

Weiss andNikitin 1998

Analyzes effects of ownership by investment funds onperformance of 125 privatized Czech firms during1993–95. Assesses these effects by measuring rela-tionship between changes in performance and incomposition of ownership at the start of privatization.Uses robust estimation techniques, in addition toOLS, since data strongly reject normality.

Ownership concentration and composition jointlyaffect performance of privatized firms. Concentrationin the hands of a large shareholder, other than an in-vestment fund or company, is associated with signifi-cant improvements for all measures of performance.Concentrated ownership by funds did not improveperformance. Preliminary post-1996 data suggestschanges in investment fund legislation may improvetheir performance.

Claessensand Djankov1999a

Studies effect of management turnover on changesin financial and operating performance of 706 pri-vatized Czech firms over the period 1993–97. Exam-ines changes in profitability and labor productivity.

Finds that the appointment of new managers is asso-ciated with significant improvements in profit marginsand labor productivity, particularly if the managers areselected by private owners. New managers appointedby the National Property Fund also improve per-formance, though not by as much.

Megginson and Netter: From State to Market 359

regarding the impact of privatization onthe performance of divested central andeastern European firms. This is espe-cially true of the five studies—Dyck(1997), Weiss and Nikitin (1998), Claes-sens and Djankov (1999b), Lubomir Li-zal, Miroslav Singer, Jan Svejnar (2000),and Frydman, Hessel, and Rapaczynski(2000)—we consider the most persua-sive due to sample size, period of cover-age and/or methodological rigor. All butone (Joel Harper 2000) of the studiesdetailed in table 6 explicitly testwhether the type of ownership struc-ture that emerges from the process is

related to post-privatization perfor-mance, and these studies documentconsistent and significant relationships.Other things equal:

i) Private ownership is associated withbetter firm-level performance thanis continued state ownership. Con-centrated private ownership is asso-ciated with greater improvementthan is diffuse ownership.

ii) Foreign ownership, where allowed,is associated with greater post-privatization performance improve-ment than is purely domestic

TABLE 6 (Cont.)

Study Sample description, study period, and methodology Summary of empirical findings and conclusions

Claessensand Djankov1999b

Examines the relationship between ownership con-centration and corporate performance for 706 pri-vatized Czech firms during the period 1992–97. Useprofitability and labor productivity as indicators ofcorporate performance.

Finds that concentrated ownership is associated withhigher profitability and labor productivity. Also findsthat foreign strategic owners and non-bank-sponsoredinvestment funds improve performance more thanbank-sponsored funds.

Frydman,Gray, HesselandRapaczynski2000

Examines whether the imposition of hard budgetconstraints is alone sufficient to improve corporateperformance in the Czech Republic, Hungary, andPoland. Employs a sample of 216 firms, split betweenstate-owned (31%), privatized (43%), and private(26%) firms.

Finds privatization alone added nearly 10 percentagepoints to the revenue growth of a firm sold to outsideowners. Most importantly, finds that the threat of hardbudget constraints for poorly performing SOEs falters,since governments are unwilling to allow these firmsto fail. The brunt of SOEs’ lower creditworthiness fallson state creditors.

Frydman,Hessel andRapaczynski2000

Examines whether privatized central European firmscontrolled by outside investors are more entrepre-neurial—in terms of ability to increase revenues—than firms controlled by insiders or the state. Studyemploys survey data from a sample of 506 manufactur-ing firms in the Czech Republic, Hungary, and Poland.

Documents that all state and privatized firms engagein similar types of restructuring, but that productrestructuring by firms owned by outside investors issignificantly more effiective, in terms of revenue gen-eration, than by firms with other types of ownership.Concludes the more entrepreneurial behavior ofoutsider-owned firms is due to incentive effects,rather than human capital effects, of privatization—specifically greater readiness to take risks.

Harper 2000 Examines the effects of privatization on the financialand operating performance of 174 firms privatized inthe first—and 380 firms divested in the second—waveof the Czech Republic’s voucher privatizations of 1992and 1994. Compares results for privatized firms tothose which remain state-owned. Employs Megginson,Nash and van Randenborgh methodology and varia-bles to measure changes.

Finds that the first wave of privatization yieldeddisappointing results. Real sales, profitability, ef-ficiency and employment all declined dramatically(and significantly). However, second wave firms ex-perienced significant increases in efficiency andprofitability and the decline in employment—thoughstill significant—was much less drastic than after firstwave (−17% vs −41%).

Lizal, Singer,and Svejnar2000

Examines the performance effects of the wave ofbreak-ups of Czechoslovak SOEs on the subsequentperformance of the master firm and the spin-offs. Theregressions use data for 373 firms in 1991 and 262firms in 1992.

There was an immediate (in 1991) positive effect onthe efficiency and profitability of small and mediumsize firms (both master and spin-offs) and negative forthe larger firms. The results for 1992 are similar butnot statistically significant.

360 Journal of Economic Literature, Vol. XXXIX (June 2001)

ownership.32 Majority ownership byoutside (non-employee) investors isassociated with significantly greaterimprovement than is any form ofinsider control.

iii) Firm-level restructuring is associ-ated with significant (sometimesdramatic) post-privatization perfor-mance improvements, and this is akey advantage of outsider control—firms controlled by non-employeeinvestors are much more likely torestructure.

iv) Most studies document that perfor-mance improves more when newmanagers are brought in to run afirm after it is privatized than whenthe original managers are retained.The precise reason for this is un-clear, though FGHR (2000) findthat the more entrepreneurial be-havior of outsider-owned firms isdue to incentive rather than humancapital effects.

v) The role of investment funds in pro-moting efficiency improvements inprivatized Czech firms is ambiguous.FGHR (1999) find selling an SOEto a domestic financial company sig-nificantly increases the growth rateof the enterprise, while Weiss andNikitin (1998) find that concen-trated ownership by investmentfunds is not associated with im-provement. Claessens and Djankov(1999b) document greater improve-ment for companies controlled by non-bank-sponsored investment fundsthan by bank-sponsored funds.

vi) There is evidence that while perfor-mance improves especially forsmaller firms, the performance im-provement declines over time. Li-zal, Singer, and Svejnar (2000) findthat for Czech firms there is an im-provement in performance amongSOE firms that are broken up,which declines over time—perhapsdue to increased competition ormanagers siphoning off profits.

vii) The impact of privatization on em-ployment is also ambiguous, primar-ily because employment falls forvirtually all firms in transitioneconomies after reforms are initi-ated. Harper (2000) documents em-ployment declines following thefirst Czech mass privatization wavein 1992, but not after the secondwave in 1994. FGHR (1999) isthe only study that explicitly exam-ined employment changes—after ac-counting for ownership structurechanges—and found that sales growfast enough in outsider-controlledfirms to offset the significantincrease in labor productivity.

viii) There is little evidence that govern-ments have been able to imposehard budget constraints on firmsthat remain state-owned after re-forms begin. FGHR (2000) findthat the threat of hard budget con-straints falters for poorly perform-ing SOEs, since governments areunwilling to allow these firms tofail. However, both FGHR andMark Schaffer (1998) show that theburden of lower SOE creditworthi-ness falls on the state (as deferredtaxes) or on state creditors, ratherthan on private creditors or suppliers.

5.2.2 Privatization in the FormerSoviet Union

Table 7 summarizes the results ofsix empirical studies that examine

32 In his analysis of the reasons why Hungary’sprivatization program has proven to be so muchmore successful than those in most other centraland eastern European countries, Peter Mihályi(2000) emphasizes the importance of selling SOEsdirectly to western transnational companies, andthus plugging them into the global trading system.Other countries stressed domestic over foreignownership, and thus missed out on the opportu-nity of using privatization as a way of attractingforeign direct investment.

Megginson and Netter: From State to Market 361

privatization programs in Russia andthe other republics of the former SovietUnion. It is very difficult to reach asimple conclusion regarding privatiza-tion’s impact in the former SovietUnion in general, and Russia in particu-lar, for four principal reasons. First, thetransition from socialism to capitalismwas much more difficult and painful inthe former Soviet Union republics thananywhere else in the world, both be-

cause these republics were under com-munist rule the longest and because thetransition to capitalism also coincidedwith dissolution of the Soviet Union.Breaking up any continental-scale na-tion would likely prove traumatic;breaking up a country that was also aneconomic system proved doubly so. Sec-ond, the contraction in output that oc-curred in the former Soviet Union after1991 was far greater than anywhere

TABLE 7SUMMARY OF EMPIRICAL STUDIES OF PRIVATIZATION IN TRANSITION ECONOMIES:

RUSSIA AND FORMER SOVIET REPUBLICS

Study Sample description, study period, and methodology Summary of empirical findings and conclusions

Barberis,Boycko,Shleifer, andTsukanova1996

Surveys 452 Russian shops sold in early 1990s tomeasure importance of alternative channels throughwhich privatization promotes restructuring.

Documents that new owners and managers raiselikelihood of value-increasing restructuring. Findsequity incentives do not improve performance; insteadpoints to importance of new human capital in eco-nomic transformation.

Earle 1998 Investigates impact of ownership structure on (labor)productivity of Russian industrial firms. Using 1994survey data, examines differential impact of insider,outsider, or state ownership on performance of 430firms, of which 86 remained 100% state-owned, 299were partially privatized, and 45 were newly created.Adjusts empirical methods to account for tendency ofinsiders to claim dominant ownership in the best firmsbeing divested.

OLS regressions show positive impact of private(relative to state) share ownership on labor pro-ductivity, primarily due to managerial ownership. Afteradjusting for selection bias, finds that only outsiderownership is significantly associated with productivityimprovements. Stresses that leaving insiders in controlof firms, while politically expedient, has negative long-term implications for restructuring of Russian in-dustry.

Earle andEstrin 1998

Using a sample similar to that used by Earle (1998),examines whether privatization, competition andhardening budget constraints enhance efficiency inRussia.

Finds a 10 percentage-point increase in private shareownership raises real sales per employee by 3–5%.Subsidies (soft budget constraints) reduce pace of re-structuring in SOEs, but the effect is small and ofteninsignificant.

Djankov1999a

Investigates relation between ownership structure andenterprise restructuring for 960 firms privatized in 6newly independent states between 1995–97. Employssurvey data collected by World Bank in late 1997 fromGeorgia, Kazakhstan, Kyrgyz Republic, Moldova,Russia, and Ukraine.

Shows that foreign ownership is positively associatedwith enterprise restructuring at high ownership levels(>30%), while managerial ownership is positively re-lated to restructuring at low (<10%) or high levels, butnegative at intermediate levels. Employee ownershipis beneficial to labor productivity at low ownershiplevels, but is otherwise insignificant.

Djankov1999b

Using same survey data as Djankov (1999a), studieseffects of different privatization modalities on re-structuring process in Georgia (92 firms) and Moldova(149 firms). Georgia employed voucher privatization,while the majority of Moldovan firms were acquiredby investment funds—and numerous others were soldto managers for cash.

Privatization through management buy-outs is posi-tively associated with enterprise restructuring, whilevoucher privatized firms do not restructure morerapidly than state-owned firms. Implies that managerswho gain ownership for free may have less incentive torestructure, as their income is not solely based onsuccess of the enterprise.

Black, Kraak-man, andTarassova2000

Surveys the history of privatization in Russia. Whilemostly descriptive, several case studies are analyzed.

Authors conclude that Russian privatization hascreated a “kleptocracy” and has essentially failed.Stresses importance of minimizing incentives for self-dealing in design of privatization programs.

362 Journal of Economic Literature, Vol. XXXIX (June 2001)

else—and there is as yet no upturn—making it very difficult to document anykind of relative performance improve-ment, or to assign causality to any im-provement that is found. Third, it seemsclear that the former Soviet Unionrepublics—especially Russia—took adecided turn for the worse economi-cally after 1997, so competently exe-cuted studies examining privatization’simpact in the same country, but at dif-ferent times, might well reach radicallydifferent conclusions. Finally, all fivestudies that examine Russia’s experi-ence rely either on survey data or anec-dotal evidence, so the “raw material”for empirical analysis is of much poorerquality here than in other regions. Forthese reasons, we believe that no trulypersuasive empirical study of privatiza-tion in the former Soviet Union has yetbeen performed, nor is one likely untilthese economies stabilize and severalyears of reliable accounting (not survey)data become available.

In spite of the difficulties (and cave-ats) spelled out above, the studies sum-marized in table 7 do yield consistentconclusions. Certainly the most impor-tant result all these studies find is thatinsider privatization has been a failurethroughout the former Soviet Union,especially in Russia, and that the con-centrated managerial ownership struc-ture that characterizes almost all priva-tized firms will likely hamper theseeconomies for many years. As describedin Bornstein (1994), John Earle (1998),Earle and Saul Estrin (1998), and Ber-nard Black, Reinier Kraakman, andAnna Tarassova (2000), Russian reform-ers considered rapid privatization to bean imperative, and for this reason theyopted for the politically expedient tech-nique of favoring incumbent managersand employees with allocations of con-trolling shareholdings during the initialmass privatization waves of 1992–93.

The investment funds created duringthis program proved ineffective, dueprimarily to insider control and poor le-gal protection of (outside) shareholdervoting rights. In spite of this, NicholasBarberis, Boycko, Shleifer, and NataliaTsukanova (1996), Earle (1998), andEarle and Estrin (1998) document thatprivatization was associated with perfor-mance improvements in firms that weredivested during the mass privatizationprogram of the early 1990s. However,all three studies, as well as Djankov(1999a,b), find that post-privatizationperformance improves the most (oronly) for firms that are outsider con-trolled, and all the studies stress the im-portance of bringing in new manage-ment. Additionally, Djankov (1999a)finds that foreign share ownership is as-sociated with significantly greater per-formance improvement than is purelydomestic ownership, and Djankov (1999b)shows that managers who actually payfor divested firms (through managementbuy-outs) improve performance more thando managers who are effectively givencontrol (through voucher schemes).

Russia provides an example of whatcan go wrong with privatizations in the1995 “loans for shares” scheme, whichtransferred control of twelve natural re-source firms to a small group of “oli-garchs” at very low prices. Black et al.(2000) argue this was a corrupt andnontransparent transfer of assets thatprecipitated widespread insider expro-priation. Further, it contributed to thepolitical unpopularity of privatization inRussia. It provides a cautionary notethat privatization is not an economicpanacea.33

33 The Czech Republic’s market collapse of1997, described in Jack Coffee (1999), and theLithuanian government’s tortuous privatization ofthe Mazheikiu Nafta refinery in early 2000, de-scribed by Val Samonis (2000), are also examplesof what can go wrong in privatization programs.

Megginson and Netter: From State to Market 363

Black et al. also argue that a poorlydesigned privatization program is worsethan none at all. However, Nellis (1999)and other commentators point out thatmany of Russia’s problems resultedfrom a collapse of central governmentalauthority and would thus not likely besolved by renationalization. Perhaps thebest long-term hope for economic revi-talization in the former Soviet Unionrepublics is the type of de novo privatedevelopment described in Havrylyshynand McGettigan (2000).

5.2.3 Summary of Evidence

Review articles by Djankov and Mur-rell (2000a,b) and a macroeconomicstudy by Jeffrey Sachs, Clifford Zinnes,and Yair Eilat (2000) examine the ef-fects of privatization in transitioneconomies. Djankov and Murrell reviewthe empirical results of studies of priva-tization in transition economies and at-tempt to synthesize the results acrossthe studies. They conclude that the evi-dence shows that: in most countries,privately owned firms perform betterthan state-owned firms, usually signifi-cantly better statistically; there is littleevidence privatization has hurt firmperformance even in Russia and otherCommonwealth of Independent States(CIS) countries; much better outcomesoccur when the new owners are concen-trated; and privatization has had a largerpositive impact in non-CIS countries,eastern and central Europe, and theBaltic states than in the CIS countries.They interpret the last result to becaused by institutional factors, includ-ing the choice of privatization method.They suggest the best empirical proxyfor how well the institutions performedwas the length of time the country hadspent under communism—the shorterthe time the better the performance ofthe institutions.

Empirically, at a macro level, Sachs,

Zinnes, and Eilat (2000) examine the rela-tionship between privatization, institu-tional reforms, and overall economicperformance (measured by change inGDP from before transition, foreign di-rect investment, and exports) in transi-tion economies. They find that changein ownership is not enough to improvemacroeconomic performance. The gainsfrom privatization come from change inownership combined with other reformssuch as institutions to address incentiveand contracting issues, hardened budgetconstraints, removal of barriers to en-try, and an effective legal and regula-tory framework. While this is a macro-economic study, the changes theyreport must come from the operationsof individual firms.

Our reading of the evidence fromtransition economies is very similar. Pri-vatization improves performance butvarious factors impact the success of theprivatization. Most important is that al-lowing incumbent managers to gain con-trol of privatized firms, through what-ever means, will yield disappointingresults. Whenever possible, firms shouldbe privatized, for cash, in as transparenta method as possible, and through anauction or sale process that is open tothe broadest possible cross-section ofpotential buyers (including foreigners).Finally, institutional factors matter, andwe discuss their implications in a latersection.

5.2.4 Economic Reform in China

China, one of the most importanttransition economies, has been vigor-ously pursuing economic reform since1978. It has dramatically increased thetotal factor productivity (Li 1997) ofChinese SOEs, largely by improving in-centives (Groves, Hong, McMillan, andNaughton 1994, 1995) and decentraliz-ing economic decision making (Yuan-zheng Cao, Yingyi Qian, and Barry

364 Journal of Economic Literature, Vol. XXXIX (June 2001)

Weingast 1999; Lawrence Lau, Qian, andGerard Roland 2000). Lau, Qian,and Roland (2000) show theoreticallyand empirically that the Chinese havesuccessfully followed a dual-track ap-proach to market liberalization, as amethod of implementing an efficientPareto-improving reform. The idea wasto continually enforce the existing plan,while liberalizing the market to makeimplicit transfers to compensate losersunder reform.

The Chinese Communist Party re-cently committed the country to a mas-sive privatization program (Lin 2000)under the slogan “seize the large, re-lease the small,” which roughly trans-lates as privatizing all but the largest300 or so SOEs. Assuming this plan iseven partially implemented, the resultwill be a privatization program of un-precedented scale. Furthermore, theWorld Trade Organization accord nego-tiated between China and the UnitedStates in November 1999 (and sub-sequently with the European Union inearly 2000) may ultimately lead toChina’s accession to the WTO. If thisoccurs, broad swathes of heretoforeprotected Chinese industry—includingtelecommunications, automobile pro-duction, and financial services—will beopened to international competition forthe first time. This process will almostcertainly increase the pressure onChina to fully privatize its industry.

On the other hand, there are reasonsto believe that China’s “privatization”program will do little to lessen thestate’s role in economic decision mak-ing, either at the macro- or micro-economic levels. For one thing, theownership structure of Chinese stockcompanies is unique. As described byXiaonian Xu and Yang Wang (1997),Tian (2000), and Lin (2000), only one-third of the stock in China’s publiclylisted former SOEs can be owned by in-

dividuals; the remaining two-thirds of acompany’s shares must be owned by thestate and by domestic (usually financial)institutions—which are invariably state-owned. So-called “A-shares” may beowned and traded only by Chinese citi-zens, while B-shares are stocks listed inShanghai or Shenzhen that may beowned and traded only by foreigners.Other shares are listed in Hong Kong(H-shares) or New York (N-shares), andthese are also restricted to foreigners.The net effect of this fractionalizationof ownership is that, even in publiclylisted former SOEs, control is never re-ally contestable, and the long-term fi-nancial performance of “privatized”Chinese companies has been quitepoor. This is particularly true for the“Red Chip” (PRC-controlled companiesincorporated and listed in Hong Kong)and H-shares sold in Hong Kong.34

These ownership restrictions could,however, be rescinded by governmentfiat at any time. Perhaps the key con-straint on privatization in China is thefact that SOEs, rather than the govern-ment itself, serve as the country’s socialsafety net. As described in Bai, Li, andWang (1997) and Lin, Cai, and Li (1998),Chinese SOEs are burdened with manysocial welfare responsibilities. Thus it isdifficult to imagine the governmentadopting a privatization program thatwould either grant these firms discre-tion over staffing levels or subject themto truly enterprise-threatening competi-tion. In sum, the long-term prognosisfor privatization in China is unclear;there is great scope for such a programto have a dramatic impact, coupled withgreat danger of social turmoil if handled(or sequenced) incorrectly.

34 We thank Cyril Lin, Samuel Huang, andGeorge Tian for helping us understand Chineselisting procedures. See http://www.csrc.gov.cn/CSRCsite/eng/elaws/elaws.htm for an English-language summary of Chinese securities laws.

Megginson and Netter: From State to Market 365

We now shift our emphasis from tran-sition economies to examining whetherinvestors who participate in share issueprivatizations have, on average, bene-fited from these investments—both ini-tially (first day) and longer term (up tofive years).

6. Do Investors Benefit fromPrivatization?

6.1 Initial Returns from SIPs

As noted earlier, governments gener-ally rely on share offerings as the bestmethod of privatizing large state-ownedenterprises, and they routinely adopthighly politicized offer terms in orderto achieve political objectives. Offeringterms that differ fundamentally fromthose observed in private-sector offer-ings, plus the very large average size ofprivatization issues, have motivatedmany researchers to examine the initialand long-term returns earned by SIP in-vestors. Table 8 summarizes the resultsof ten studies examining initial returns.Most of these studies evaluate whetherinvestors who purchase privatizationinitial public offerings (PIPOs) at theoffering price, and then sell theseshares on the first day of open markettrading, earn returns that are signifi-cantly different from zero. These stud-ies test whether PIPOs are “under-priced.” A few also test whether PIPOsyield initial returns that are materiallydifferent from the significantly positivefirst-day returns earned by investors inprivate-sector IPOs, as documented in avast number of articles using both U.S.and international data. The U.S. marketexperience is summarized in Roger Ib-botson, Jody Sindelar, and Jay Ritter(1994), and international IPO under-pricing studies are surveyed in TimLoughran, Ritter, and Kristian Rydqvist(1994).

Five of the studies in table 8 examine

PIPO returns from individual countries.All five studies document significant,often massive, average levels of under-pricing, ranging from 39.6 percent forthe forty British PIPOs studied by Men-yah and Paudyal (1996) to 940 percentfor the 308 Chinese PIPOs Class A is-sues (domestic issuance) examined byDongwei Su and Belton Fleisher (1999).Menyah and Paudyal, and Paudyal, B.Saadouni, and R. Briston (1998) findthat U.K. and Malaysian PIPOs are sig-nificantly more underpriced than theirprivate-sector counterparts, and WolfgangAusenegg (2000) finds the same resultfor Polish PIPOs. Hungarian PIPOs arealso more underpriced than privateIPOs, but the difference is not signifi-cant (Ranko Jelic and Richard Briston2000b). Since there are as yet few trulycomparable private-sector IPOs inChina, Su and Fleisher cannot testwhether private offerings also have theincredible underpricing they documentfor PIPOs. They do find that Class Bshares, issued internationally, are muchless underpriced (37 percent mean ini-tial return). Unlike almost any compara-ble group of IPOs, over 90 percent ofChinese PIPOs do in fact execute sea-soned equity offerings within a shorttime after the PIPO. Further, the prob-ability of a seasoned offer occurring ispositively related to the level of the ini-tial offer underpricing, which is consis-tent with various signalling models,including Ivo Welch (1989).

The other five studies in table 8 ex-amine multinational samples of PIPOs,generally using offering data from Pri-vatisation International and stock re-turns from Datastream. The number ofcountries studied ranges from eight inDewenter and Malatesta (1997) to 61 inAlexander Ljungqvist, Tim Jenkinson,and William Wilhelm (2000), thoughthe studies’ main results are similar. Allthese studies document economically

366 Journal of Economic Literature, Vol. XXXIX (June 2001)

TABLE 8SUMMARY OF EMPIRICAL STUDIES EXAMINING INITIAL RETURNS TO INVESTORS IN SHARE ISSUE PRIVATIZATION

(Return earned by investors who buy shares in SIPs at the offer price and sell the shares immediately after trading begins.)

Study Sample description, study period, and methodology Summary of empirical findings and conclusions

Menyah andPaudyal 1996

Examines initial and long-term returns for 40 UKprivatization IPOs (PIPOs) and 75 private-sector IPOson London Stock Exchange between 1981–91.

PIPOs offer a market-adjusted initial return of 39.6%,compared to private sector IPO initial return of 3.5%.Regression analysis explains up to 64% of variation inPIPO initial returns.

DewenterandMalatesta1997

Tests whether privatization IPOs (PIPOs) are more orless underpriced than private sector IPOs in 8 coun-tries. Compares actual initial returns for 109 companiesfrom Canada, France, Hungary, Japan, Malaysia, Poland,Thailand and UK with national average initial returnsreported in Loughran, Ritter, and Rydqvist 1994.

Mixed results. Initial returns to privatization issues arehigher than to private sector IPOs in unregulated in-dustries and in UK. Privatization IPOs are lower thanprivate offers in Canada and Malaysia; but there is nosystematic tendency to underprice PIPOs on the partof all governments.

Huang andLevich 1998

Studies offering terms and initial returns to investorsin 507 privatization share offerings from 39 countriesduring 1979–96; tests alternative explanations forobserved underpricing.

Average initial returns of 32.2% for PIPOs and 7.17%for seasoned privatization offerings. SIPs from non-OECD countries are more underpriced than OECDoffers, but there is no evidence PIPOs are under-priced more than private IPOs.

Paudyal,Saadouni,and Briston1998

Examines initial and long-term returns offered toinvestors in 18 PIPOs and 77 private sector IPOs inMalaysia from 1984–95. Provides details of offeringterms and share allocation patterns.

Malaysia PIPOs offer market-adjusted initial returnsof 103.5% (median 79.9%), significantly greater thanthe private sector IPO initial returns of 52.5% (29.4%).

Jones,Megginson,Nash, andNetter 1999

Examines how political and economic factorsinfluence initial returns, and share and controlallocation patterns, for a sample of 630 SIPs from 59countries during 1977–97.

Governments deliberately underprice both PIPOs (mean34.1%, median 12.4%) and seasoned SIPs (9.4% and3.3%). Share and control allocation patterns are bestexplained by political factors. Support predictions ofBiais and Perotti (2000) theoretical model.

Su andFleisher 1999

Studies cross-sectional pattern of underpricing of 308Chinese PIPOs from 1987–95. Tests whetherobserved underpricing for domestic shares can beexplained using a signalling model.

Massive underpricing, with average initial return of940% on A shares (issued domestically). Findingsconsistent with a signalling model, since 91% of allfirms subsequently execute seasoned equity offerings.Less underpricing for B shares (international).

Jelic andBriston2000b

Examines initial and long-term returns for 25 PIPOsand 24 other IPOs in Hungary during 1990–98.

PIPOs are much larger and have higher market-adjusted initial returns than other IPOs (44% meanand 9% median vs 40% and 5%, respectively), butreturn differences are insignificant.

Jelic andBriston 2000a

Examines initial and long-term returns for 55 PIPOsand 110 other IPOs in Poland during 1990–98.

Using first-day opening prices (not offer prices), findssmall, significant positive mean abnormal initial re-turns (1.16%) for PIPOs and insignificant mean abnor-mal initial returns (0.22%) for other IPOs. The differ-ence is insignificant.

Ausenegg2000

Examines initial and long-term returns for 52 PIPOsand 107 other IPOs in Poland during 1990–98.

Significantly positive initial abnormal return for PIPOs(60.4% mean, 19.8% median) and other IPOs (19.8%and 12.9%), though difference is insignificant. With-out Bank Slaski, mean PIPO initial return cut roughlyin half.

Choi andNam 2000

Compares initial returns of 185 PIPOs from 30countries during 1981–97 to those of private sectorIPOs from same countries, using mean national initialreturns reported in Loughran, Ritter, and Rydqvist1994.

PIPOs tend to be more underpriced than private sec-tor IPOs (mean 31% vs 24.6%), and underpricing forPIPOs is positively related to the stake sold and todegree of uncertainty in ex-ante value of newly pri-vatized firms.

Ljungqvist,Jenkinson,and Wilhelm2000

Analyzes direct and indirect costs (associated withunderpricing) of 2,051 IPOs, including 185 PIPOs, in61 non-US markets during 1992–99. Primarily aprivate-sector, underwriting study.

PIPOs are significantly more underpriced (by about 9percentage points) than are private-sector IPOs, andunderwriter spreads are a significant 61 basis pointslower.

and statistically significant underpricingof PIPOs, averaging about 30 percent inthe large-sample studies. The two thatexamine seasoned SIPs (Huang andLevich 1998; and Jones, Megginson,Nash, and Netter 1999) find these aresignificantly underpriced as well, thoughmuch less so than are PIPOs. Four ofthese studies—Dewenter and Malatesta(1997), Huang and Levich (1998),Seung-Doo Choi and Sang-Koo Nam(2000), and Ljungqvist et al. (2000)—also test whether PIPOs are signifi-cantly more underpriced than private-sector IPOs. The first three studies findno systematic evidence that PIPOs aresignificantly more or less underpricedthan private IPOs; instead all three sug-gest that results vary by country. How-ever, the Ljungqvist et al. study per-forms the most convincing analysis ofthe relative underpricing of IPOs andPIPOs, since they use regression meth-odology and a privatization dummyvariable to examine underpricing for asample of 2,051 IPOs—including 185PIPOs—from 61 non-U.S. markets.They document that PIPOs are signifi-cantly more underpriced (by about 9percentage points) than are private sec-tor IPOs. They also find that the under-writing spreads on PIPOs are signifi-cantly lower (by a mean 61 basis points)than on IPOs.

The principal objective of the Joneset al. (1999) study differs from the oth-ers in that it tests whether governmentissuers are attempting to maximize SIPoffering proceeds or are instead tryingto achieve multiple political and eco-nomic objectives, even at the cost ofrevenue maximization. Jones et al.(1999) test the underpricing models ofPerotti (1995) and Biais and Perotti(2000). Both models predict that gov-ernments that are ideologically commit-ted to privatization and economic re-form will deliberately underprice SIPs

and will privatize in stages, to signaltheir commitment to protecting inves-tor property rights. “Populist” govern-ments pursuing privatization strictly asa means of raising revenue will be un-willing to underprice as much as willcommitted governments. Populist gov-ernments will also try to sell largerstakes in SOEs. Jones et al. (1999) findthat initial returns (underpricing) aresignificantly positively related to thefraction of the firm’s capital sold and tothe degree of income inequality (Ginicoefficient) in a country. They also findthat initial returns are negatively re-lated to the level of government spend-ing as a fraction of GDP (a proxy for howsocialist a society is) and to a dummyvariable indicating that more than 50percent of a company’s stock is being sold.Collectively, these findings stronglysupport the predictions of Perotti (1995)and Biais and Perotti (2000).

6.2 Long-Run Returns from SIPs

Since the seminal article by Jay Ritter(1991), financial economists have paidclose attention to estimating the long-run returns earned by investors whopurchase unseasoned and seasonedissues. Most of these papers find sig-nificantly negative long-term returns,whether they examine U.S. offeringsor international stock issues, though afew studies document insignificantlypositive long-term performance.35

There is a major debate in the em-pirical finance literature on methodo-logical issues in estimating long-run

35 Early long-run return studies, using both U.S.and international data, are summarized inLoughran, Ritter, and Rydqvist (1994). Later stud-ies employing U.S. data and finding negative long-run returns include Tim Loughran and Ritter(1995, 1997), Katherine Spiess and John Affleck-Graves (1995), and Richard Carter, FrederickDark, and Ajai Singh (1998). Only a few U.S. stud-ies, including Alon Brav and Paul Gompers (1997),find (insignificantly) positive long-term returns.

368 Journal of Economic Literature, Vol. XXXIX (June 2001)

returns. This is not surprising, sincefindings of significant negative (or posi-tive) long-run returns can be inter-preted as evidence contradicting theefficient market hypothesis, a funda-mental concept in finance. The debatecenters on how to calculate long-run re-turns and how to construct test statis-tics. For example, Mark Mitchell andErik Stafford (2000) argue that mostcorporate actions are not randomevents. They contend that after control-ling for cross-correlation of abnormalreturns, most statistical evidence of ab-normal performance disappears. JohnLyon, Brad Barber, and Chih-Ling Tsai(1999), drawing on the work of S. P.Kothari and Jerold Warner (1997) andBarber and Lyon (1997), note five rea-sons for misspecification in test statis-tics designed to detect long-run re-turns. There are three sources ofbias—a new listing bias, a rebalancingbias, and a skewness bias—as well ascross-sectional dependence in sampleobservations and a poorly specifiedasset-pricing model. Lyon, Barber, andTsai, among others, suggest severalmethods to control for misspecification,but there is no one correct method.They conclude that the “analysis oflong-run returns is treacherous.” LindaCanina, Roni Michaely, Richard Thaler,and Kent Womack (1998) present an-other approach to dealing with long-runreturns, and Fama (1998) argues badmodel problems are “unavoidable . . .and more serious in tests on long-termreturns.” Two other papers that do anexcellent job of analyzing the problemswith estimating long-run returns areAlon Brav, Christopher Geczy, and PaulGompers (2000), and B. Espen Eckbo,Ronald Masulis, and Øyvind Norli(2000).

Since the methodological problemsidentified with estimates of long-run re-turns have not been resolved for U.S.

firms, they have not been resolved forprivatizations that are subject to the ad-ditional problems of scarce data and thelack of liquid markets. Nevertheless,the fact that most of the studies of long-run returns following privatizations—using different methodologies and fo-cusing on different countries—findsimilar results lessens some of themethodological concerns.

We discuss fifteen studies that exam-ine the returns earned by investors whobuy and hold privatization share issues,and the number of such studies appearsto be growing rapidly. The papers aresummarized in table 9. Eight of thesefocus on either a single country or a sin-gle market for issues, and the otherseven examine multinational samples.Mario Levis (1993) and Menyah,Paudyal, and Inganyete (1995) examinethe British experience, and both docu-ment significantly positive long-run ab-normal returns for SIP investors. How-ever, Reena Aggarwal, Ricardo Leal,and Leonardo Hernandez (1993) findthe opposite result for their sample ofnine Chilean SIPs. Ranko Jelic andRichard Briston (2000b) find that 25Hungarian PIPOs yield large but insig-nificantly positive long-run returns(peaking at 21.3 percent in month 15),though they do find that these cumula-tive returns are significantly higherthan the highly negative returns (reach-ing –70 percent by month 30) earned on24 private-sector IPOs. Jelic and Bris-ton (2000a) document significantly posi-tive one-, three-, and five-year excessreturns for Polish PIPOs, but Ausenegg(2000) finds insignificant long-term re-turns for essentially the same sample.Given the differing estimation method-ologies employed in these two studies,it is not clear whether Polish PIPOs earnsignificantly higher long-run returns thanIPOs. Stephen Foerster and G. AndrewKarolyi (2000) find insignificant long-run

Megginson and Netter: From State to Market 369

TABLE 9EMPIRICAL STUDIES OF LONG-RUN RETURNS TO INVESTORS IN SHARE ISSUE PRIVATIZATIONS

1–5 year returns earned by investors who buy and hold offerings. Unless otherwise noted, long-run returnexcludes first-day return to issue date.

Study Sample description, study period, and methodology Summary of empirical findings and conclusions

Levis 1993 Examines long-run return to 806 British IPOs from1980–88. Sample includes 12 PIPOs, accounting for76% of total IPO value.

Private sector IPOs underperformed the market by+10% over 3 years, PIPOs outperformed the marketby +15%.

Aggarwal,Leal andHernandez1993

Examines long-run (1-year) returns for Latin Ameri-can IPOs, including 9 Chilean PIPOs from 1982–90.

Using returns from offer price, finds significantnegative 1-year market-adjusted returns for PIPOsaveraging –29.9% (median −32.4%) vs −9.8%(−23.0%) for private sector IPOs.

Menyah,Paudyal andInganyete1995

Examines initial and long-term returns for 40 BritishPIPOs and 75 private sector IPOs executed on Lon-don Stock Exchange 1981–91.

Significant positive 33% market-adjusted 400-day (80-week) return for PIPO vs insignificant 3.5% return forprivate sector IPOs.

Davidson1998

Studies 1, 3, 5, and 10-year market adjusted returnsfor SIPs from 5 European countries (Austria, France,Italy, Spain, UK) through March 1997.

After long period of underperformance, averaging 1–1.5% per year, finds SIPs outperformed Europeanmarket averages in previous 12 months.

Foerster andKarolyi 2000

Examines long-run return for 333 non-US firms thatlist stock on US markets in the form of ADRs in 1982–96. Compares returns for 77 SIPs (38 IPOs, 39 sea-soned offers) with private offers.

Insignificantly positive 4.1% 3-year abnormal returnsfor SIPs vs insignificantly negative returns of −1.7%for full sample.

Paudyal,Saadouni,and Briston1998

Examines initial and long-term returns offered to in-vestors in 18 PIPOs and 77 private sector IPOs in Ma-laysia 1984–95. Provides details of offering terms andshare allocation patterns.

PIPOs and private sector IPOs yield normal returns(insignificantly different from overall market) over 1,3, and 5-year holding periods.

Boubakriand Cosset2000

Evaluates long-term returns to investors in 120 SIPsfrom 26 developing countries during 1982–95.

Significant 3-year raw returns (112% mean, 30% me-dian), but insignificant mean (37–46%) and median(−7% to 13%) market-adjusted returns, due to weight-ing of SIPs in stock market indices. Significant posi-tive long-run returns after adjusting for impact of SIPsize on index.

Jelic andBriston 2000

Examines initial and long-term returns for 25 PIPOsand 24 other IPOs in Hungary during 1990–98.

PIPOs yield insignificantly positive market-adjustedreturns over 1, 2, and 3-year holding periods, peakingat 21.3% in month 15; private-sector IPOs yield signifi-cantly negative returns.

Jelic andBriston2000b

Examines initial and long-term returns for 55 PIPOsand 110 other IPOs in Poland during 1990–98.

PIPO investors earn positive 1, 3, and 5-year market-adjusted returns; other IPO investors earn negativereturns. Significant differences for most holding pe-riods.

Ausenegg1000

Examines initial and long-term returns for 52 PIPOsand 107 other IPOs in Poland during 1990–99.

PIPO and private-sector IPO investors earn negative—often significant—abnormal returns over 1, 3, and 5-year holding periods.

Perotti andOijen 2000

Develops a theoretical model suggesting that long-runreturns to investors in developing-country SIPs willearn excess returns if/when political risk is resolved.Tests the model using data from 22 countries with ac-tive privatization programs during 1988–95.

Their proxy for political risk declines by an annual av-erage of 3.6% during the course of a privatization pro-gram, and stock markets develop rapidly. Decline inrisk leads to positive excess returns for SIPs of about6% per year.

Choi, Nam,and Ryu 2000

Computes buy-and-hold returns of 204 PIPOs from 37countries in 1977–97.

Significantly positive market-adjusted returns to SIPsover 1, 3, and 5-year holding periods.

Megginson,Nash, Netter,and Schwartz2000

Examines long-run (1, 3, and 5-yr.) returns for 158PIPOs from 33 countries from 1981–97. Computeslocal-currency and $ returns vs national and interna-tional indices, and vs matching firms.

Economically and statistically positive holding-periodreturns in both local currency and $, and vs all marketindices. 5-year excess returns exceeding 80% arefound for most comparisons.

370 Journal of Economic Literature, Vol. XXXIX (June 2001)

returns for privatization stocks listing inthe United States in the form of Ameri-can depository receipts (ADRs) com-pared to local benchmarks. The returnsare significantly negative compared toU.S. benchmarks. Paudyal, Saadouni,and Briston (1998) find that investorsearn insignificant long-term returns oneighteen Malaysian PIPOs, as well as on77 private-sector IPOs.

Two of the multinational studies de-scribed in table 9 focus on long-run re-turns earned by investors in SIPs fromdeveloping countries. A third examinesonly western European offerings. Nar-jess Boubakri and Jean-Claude Cosset(2000) study returns from 120 SIPsfrom 26 developing countries, whileEnrico Perotti and Pieter Oijen (2000)develop and test a model of long-termreturns using data from twenty develop-ing nations. Both studies documentlarge, highly significant long-run re-turns, though the mean 112 percentthree-year return found by Boubakriand Cosset is not significant once thereturns from national markets over thecorresponding time periods are sub-tracted (the absolute returns are con-verted into market-adjusted or excessreturns). This is primarily due to the ex-

tremely large weightings that SIPsthemselves have in most developing-country national stock market indices.Once these size biases are accountedfor, SIPs significantly outperform mostnational market indices. Perotti andOijen document significantly positivemarket-adjusted returns, and argue thatthis results from a progressive resolu-tion of political risk as governments re-frain from expropriating investors’wealth in privatized firms, which hadbeen feared. Their proxy for politicalrisk declines by an average of 3.6 per-cent annually during the course of a pri-vatization program, and this leads topositive excess returns for SIPs of about6 percent per year. Richard Davidson(1998) documents that large EuropeanSIPs began to outperform market indi-ces in five countries during the mid-1990s. However, these SIPs did so onlyafter an extended period of sub-parperformance.

The remaining four long-run returnstudies employ multinational samplesthat cover a large number of countriesand regions. For this reason, and be-cause all the studies are recent enoughto employ state-of-the-art techniquesfor computing net-of-market returns,

TABLE 9 (Cont.)

Study Sample description, study period, and methodology Summary of empirical findings and conclusions

Dewenterand Mala-testa 2001

Examines long-run returns to investors in 102 SIPsfrom developed and developing countries over 1981–94. Examines long-run stock returns of privatized firmsand compares relative performance of large sample(1,500 firm-years) of state- and privately owned firmsin 1975, 1985, and 1995.

Significant positive long-term (1–5 years) abnormalstock returns, mostly concentrated in Hungary, Po-land, and UK.

Boardmanand Laurin2000

Examines factors influencing long-run returns of 99SIPs in 1980–95. Tests effect of relative size, fractionretained (by government), presence of golden share,initial return, and timing on 3-year buy-and-holdreturns. Examines whether UK utility SIPs earned“excessive” returns.

Significant positive abnormal returns to all SIPs over 1(9.2%), 2 (13.5%) and 3-year (37.4%) holding periods.UK SIPs are higher than non-UK issues, and UKutilities have highest returns (60.6% 3-year excessreturns), but 3-year non-UK SIP returns also sig-nificant. Excess returns are positively related tofraction retained and initial period return; negativelyrelated to relative size and presence of golden share.

Megginson and Netter: From State to Market 371

we consider these the most persuasiveevidence on long-term excess returnsearned by SIP investors. Megginson,Nash, Netter, and Adam Schwartz(2000) examine the long-run buy-and-hold returns earned by domestic, inter-national, and U.S. investors who pur-chase shares at the first open-marketprice in 158 SIPs from 33 countriesduring the period 1981–97. They useseveral benchmarks and compute one,three, and five-year local currency andU.S. dollar net returns with respect todomestic, international, U.S. market in-dices, and industry-matched compari-son samples. They find statistically sig-nificant positive net returns for the 158unseasoned SIPs for all holding periodsand versus all benchmarks. Boardmanand Laurin (2000), Choi, Nam, andGui-Youl Ryu (2000), and Dewenter andMalatesta (2000) find similar results.All four studies document significantlypositive market-adjusted returns overholding periods of up to five years. Ingeneral, British privatizations yieldhigher long-run returns than do non-U.K. initial and seasoned SIPs, andBritish utilities yield the highest returnsamong the U.K. offerings. However, thenet return is significantly positive formost non-U.K. subsamples as well.

These studies, and those cited ear-lier, support the conclusion that the av-erage long-term, market-adjusted re-turn earned by international investorsin share issue privatizations is economi-cally and significantly positive. Apartfrom Perotti and Oijen, however, few ofthese studies can offer any convincingexplanation of precisely why SIP issuesoutperform over time, and isolating oneor more specific cause-and-effect rela-tionships is likely to prove extremelydifficult. Most likely, these excess re-turns result from a gradual resolution ofuncertainty on the part of investors re-garding both the microeconomic suc-

cess of privatization programs and theability of governments to resist thetemptation to expropriate shareholderwealth in privatized firms through di-rect intervention or targeted regulationor taxation. If so, an important implica-tion is that returns on SIPs are likely tobe much lower in the future than theyhave been historically, since investorswill no longer demand a political riskpremium to purchase shares. The deter-minants of the long-run returns will bean interesting source of future research.

7. Privatization’s Impact on FinancialMarket Development

7.1 Stock Market Capitalization andTrading

There is no doubt that privatizationhas had a major impact on capital mar-kets. Table 10 describes the growth inthe total market capitalization, and inthe value of shares traded, on theworld’s stock exchanges from 1983 to1999. This was a period of rapid growthin the capitalization of markets in everycountry except Japan, which suffered afour-year, 70 percent decline in totalmarket capitalization after reaching avalue of $4.4 trillion in 1989. At year-end 1999, Japan’s market was eighttimes as valuable in dollar terms (andless than four times as valuable in yenterms) as it was in 1983. By contrast,total world market capitalization in-creased over tenfold (to $35.0 trillion)between 1983 and 1999, and the totalcapitalization of the U.S. market in-creased almost ninefold (from $1.9 tril-lion to $16.6 trillion) over the same pe-riod. The growth in markets outside theUnited States was even greater. It is alsoin these markets that privatization’simpact has been greatest, since therehave been only two significant SIPs inthe United States in the modern era(Conrail in 1987 and U.S. Enrichment

372 Journal of Economic Literature, Vol. XXXIX (June 2001)

Corporation in 1999). Between 1983and 1999, the total capitalization ofnon-U.S. stock markets increased from$1.49 trillion to $18.36 trillion. The totalmarket capitalization of developingcountry stock exchanges increased by26 times during these sixteen years,even after declining significantly from1997’s peak value of $2.5 trillion to $2.2trillion in 1999.

Though the rise in market capitaliza-tion has been impressive, trading vol-umes have increased even more. Thetotal value of shares traded worldwidebetween 1983 and 1999 rose from $1.2trillion to more than $37.5 trillion. Asbefore, non-U.S. markets experiencedthe greatest increases. The value ofshares traded on markets in developingcountries rose from $25 billion in 1983to more than $2.3 trillion in 1999. Thisrise in market liquidity was probably

due in large part to the increasingpopularity of “emerging market” invest-ing among western investors, particu-larly institutional investors such aspension and mutual funds.

What role has privatization played inthis remarkable growth in market capi-talization and trading volume? At theend of 1983, the total market capitaliza-tion of the handful of British, Chilean,and Singaporean firms that had beenprivatized was less than $50 billion. Bythe middle of 2000, the 152 privatizedfirms listed in either the Business Week“Global 1000” ranking of the most valu-able companies in developed-nationstock markets or the Business Week“Top 200 Emerging Market Compa-nies” ranking had a total market capi-talization of $3.31 trillion. This equalsapproximately 13 percent of the com-bined market capitalization of the firms

TABLE 10THE GROWTH OF WORLD STOCK MARKET CAPITALIZATION AND TRADING VOLUME, 1983–99

Aggregate market capitalization and trading volume in $US millions

1983 1986 1989 1992 1995 1998 1999

Market Capitalizationa

Developed countries 3,301,117 6,378,234 10,957,463 9,921,841 15,842,152 24,530,692 32,820,474United States 1,898,063 2,636,598 3,505,686 4,485,040 6,857,622 12,926,177 16,642,462Japan 565,164 1,841,785 4,392,597 2,399,004 3,667,292 2,495,757 4,554,886United Kingdom 225,800 439,500 826,598 927,129 1,407,737 2,372,738 2,855,351

Developing countries 83,222 135,056 755,210 1,000,014 1,939,919 1,908,258 2,184,899Total World 3,384,339 6,513,290 11,712,673 10,921,855 17,782,071 26,519,773 35,005,373

World, ex. US 1,486,276 3,876,692 8,206,987 6,436,815 10,924,449 13,593,596 18,362,911US as % of world 56.1% 40.5% 29.9% 41.1% 38.6% 48.7% 47.5%

Trading Volumeb

Developed countries 1,202,546 3,495,708 6,297,069 4,151,573 9,169,761 20,917,462 35,187,632United States 797,123 1,795,998 2,015,544 2,081,658 5,108,591 13,148,480 19,993,439Japan 230,906 1,145,615 2,800,695 635,261 1,231,552 948,522 1,891,654United Kingdom 42,544 132,912 320,268 382,996 510,131 1,167,382 3,399,381

Developing countries 25,215 77,972 1,170,928 631,277 1,046,546 1,956,858 2,320,891Total world 1,227,761 3,573,680 7,467,997 4,782,850 10,216,307 22,874,320 37,508,523

World, ex. US 430,638 1,777,682 5,452,453 2,701,192 5,107,716 9,725,840 17,515,084US as % of world 64.9% 50.3% 27.0% 43.5% 50.0% 57.5% 53.3%

Sources: Data sources: 1983–98, the World Bank’s Emerging Markets Fact Book (various issues); 1999 data from Statistics section ofthe International Federation of Stock Exchange’s website (www.fibv.com).a Year-end values, translated from local currencies into US$ at the contemporaneous exchange rate.b Total value of all trades executed during the year.

Megginson and Netter: From State to Market 373

on the two lists, and is more than 27percent of the non-U.S. total. (Americanfirms accounted for 484 of the “Global1000” firms, and $13.1 trillion of the $23.9trillion “Global 1000” total capitalization.)

An examination of the historical evo-lution of non-U.S. stock markets since1980 suggests that large SIPs played akey role in the growth of capital mar-

kets almost everywhere, especially be-cause they are generally among thelargest firms in national markets. Usingthe Business Week 2000 “Global 1000”and “Top 200” data, table 11 details thetotal market value and relative size ofthe world’s 25 most valuable privatizedfirms. Columns 1 and 2 give the com-pany names and domicile countries.

TABLE 11MARKET VALUES OF 25 LARGEST PUBLICLY TRADED PRIVATIZED FIRMS

Company CountryGlobal 1000

RankCountry

Rank

MarketValue US

$milc

Market Valueas % of

NationalMarket

Capitalization

NTT DoCoMo Japan 8 1 247,237 5.4BP Amoco UK 12 2 207,506 7.3Nippon Telegraph & Telephone Japan 15 2 189,156 4.2Deutsche Telekom Germany 16 1 187,247 13.1France Telecom France 25 1 148,711 9.9TotalFinaElf France 33 2 116,318 7.7China Telecom China 42a 1 102,464 16.8b

British Telecom UK 45 4 93,701 3.3Telecom Italia Italy 54 1 85,258 11.7TIM (Telecom Italia Mobiliare) Italy 60 2 75,917 10.4Telefonica Spain 72 1 66,571 15.4ING Groep Netherlands 92 3 57,474 8.3ENEL Italy 98 3 53,418 7.3STMicroelectronics France 101 6 51,324 3.4Telstra Australia 108 1 49,915 11.7Cable & Wireless UK 121 10 45,941 1.6Banco Bilbao Vizcaya Argentaria Spain 127 2 43,359 10.1ENI Italy 128 4 43,058 5.9BNP Paribas France 139 10 40,390 2.7Sonera Finland 147 2 37,199 10.6Telefonos de Mexico Mexico 151a 1 36,383 23.6CGNU UK 164 14 33,957 1.2SK Telecom Korea 186 2 30,388 9.9Cable & Wireless HKT Hong Kong 195 2 27,780 4.6Swisscom Switzerland 206 8 25,732 3.7

Source: Data are from Morgan Stanley Capital International, as reported in “The Business Week Global 1000,”Business Week (July 10, 2000). Global 1000 Rank refers to a company’s global ranking based on market valuation,while Country Rank refers to its relative position among firms from its country on the Global 1000 List.a These firms are from a companion “Top 200 Emerging-Market Companies” ranking in the same Business Weekissue, and they are given the rankings they would have if this list was included in the “Global 1000” list.b Expressed as a percentage of the Hong Kong market’s total capitalization.c Stock market value, total sales, and total profits (in US $mil. translated at contemporaneous exchange rate) of the25 most valuable publicly traded privatized firms as of May 31, 2000.

374 Journal of Economic Literature, Vol. XXXIX (June 2001)

Column 3 shows each firm’s ranking inthe “Global 1000” list (firms from the“Top 200 Emerging Markets” list aregiven the ranking they would have if in-cluded in the “Global 1000” ranking).Column 4 gives the firm’s rankingwithin its home market, and column 5lists the firm’s total market capitaliza-tion. Column 6 expresses the singlefirm’s market capitalization as a per-centage of the entire national market’syear-end 1999 capitalization.

Table 11 plus data reported in MariaBoutchkova and Megginson (2000) re-veal the relative importance of SIPs inmost non-U.S. stock exchanges. Priva-tized firms are the most valuable compa-nies in Japan, Germany, France, Italy,Spain, Australia, Mexico, Singapore,China, Denmark, New Zealand, Portugal,Russia, Argentina, Brazil, Greece, Ma-laysia, Poland, the Czech Republic,Hungary, Turkey, Indonesia, Egypt,and Peru. They are the second-mostvaluable firms in many other countries,including Britain, Finland, Hong Kong,Korea, Chile, and the Philippines. Pri-vatized companies are the first and second-most valuable companies in Japan,France, Spain, Argentina, and Indone-sia, and they occupy the three top slotsin Italy, Portugal, Russia, and Greece.Table 11 shows that the largest priva-tized firms often account for sizeablefractions of the total capitalization ofnational stock markets, even in ad-vanced countries such as Germany(13.1 percent), Italy (11.7 percent),Spain (15.4 percent), and Australia(11.7 percent). In developing countriessuch as Korea (9.9 percent) and Mexico(23.6 percent), individual privatizedfirms also account for large fractionsof the total market capitalization.

Another way to measure the impactof privatized firms on capital market de-velopment is to see how important SIPshave been as security offerings, and

here the impact is even greater. As ta-ble 12 shows, the ten largest, and thirtyof the 35 largest, share offerings in his-tory have been privatizations. Ten SIPshave been larger than the biggest U.S.share offering, the $10.6 billion AT&TWireless tracking stock offering inApril 2000. Jones, Megginson, Nash,and Netter (1999) show that, between1984 and 1997, 112 SIPs raised at least$1 billion, a stock offering size rarelyobserved in the United States. Twenty-five SIPs have raised more than $7billion, a feat no private-sector issuerachieved prior to April 2000, and gov-ernments have raised a total of morethan $700 billion through some 750public share offerings since 1977.Outside of the entire U.S. corporatesector, this is an unprecedented vol-ume of common equity issuance, andit has fundamentally changed the na-ture of global stock market trading andinvestment.

Why should we care about privatiza-tion’s impact on the development ofcapital markets? Obviously, new sharelistings can directly create some netnew wealth and a handful of new (albeitwell-paying) jobs, but the principaleconomic payoff from increasingly effi-cient and liquid capital markets comesfrom the financing opportunities andmonitoring possibilities they provide.Several studies (Ross Levine 1997; AsliDemirgüç-Kunt and Yojislav Maksimovic1998; Levine and Sara Zervos 1998; Ra-jan and Luigi Zingales 1998; AvandharSubrahmanyam and Sheridan Titman1999; Thorsten Beck, Levine, and Nor-man Loayza 2000; Geert Bekaert andCampbell Harvey 2000; Jeffrey Wurgler2000; and Peter Blair Henry 2000a,b)document that efficient capital marketspromote economic growth and allowindividual firms to fund investmentopportunities they otherwise wouldhave to forgo. Therefore, privatization

Megginson and Netter: From State to Market 375

deserves credit for whatever direct roleit has played in promoting stock marketdevelopment (through new share offer-ings), and for the indirect role it hasplayed in bond market development.

This catalytic role can be assumed be-cause several of the aforementionedstudies find development of one marketalso promotes development of relatedmarkets.

TABLE 12WORLD’S LARGEST SHARE OFFERING

Share offerings raising over $5 billion as of August 15, 2000. Offers reported in nominal amounts (notinflation-adjusted), and translated into millions of US dollars using the contemporaneous exchange rate.

Date Company Country Amount ($mil) IPOa/SEOb

Nov 87 Nippon Telegraph & Telephone Japan 40,260 SEOOct 88 Nippon Telegraph & Telephone Japan 22,400 SEONov 99 ENEL Italy 18,900 IPOOct 98 NTT DoCoMo Japan 18,000 IPOOct 97 Telecom Italia Italy 15,500 SEOFeb 87 Nippon Telegraph & Telephone Japan 15,097 IPONov 99 Nippon Telegraph & Telephone Japan 15,000 SEOJun 00 Deutsche Telekom Germany 14,760 SEONov 96 Deutsche Telekom Germany 13,300 IPOOct 87 British Petroleum United Kingdom 12,430 SEOApr 00 ATT Wireless (tracking stock)c United States 10,600 IPONov 98 France Telecom France 10,500 SEONov 97 Telstra Australia 10,530 IPOOct 99 Telstra Australia 10,400 SEOJun 99 Deutsche Telekom Germany 10,200 SEODec 90 Regional Electricity Companiesd United Kingdom 9,995 IPODec 91 British Telecom United Kingdom 9,927 SEOJun 00 Telia Sweden 8,800 IPODec 89 UK Water Authoritiesd United Kingdom 8,679 IPODec 86 British Gas United Kingdom 8,012 IPOJun 98 Endesa Spain 8,000 SEOJul 97 ENI Italy 7,800 SEOApr 00 Oracle Japanc Japan 7,500 IPOJul 93 British Telecom United Kingdom 7,360 SEOOct 93 Japan Railroad East Japan 7,312 IPODec 98 Nippon Telegraph & Telephone Japan 7,300 SEOOct 97 France Telecom France 7,080 IPOJul 99 Credit Lyonnais France 6,960 IPOFeb 94 Elf Acquitaine France 6,823 SEOJun 97 Halifax Building Societyc United Kingdom 6,813 IPOJun 98 ENI Italy 6,740 SEOMay 94 Autoliv Sverigec Sweden 5,818 IPOOct 96 ENI Italy 5,864 SEOOct 98 Swisscom Switzerland 5,600 IPOJul 99 United Parcel Servicec USA 5,500 IPO

Amounts reported for share issue privatization (SIP) offers are as described in the Financial Times at the time of theissue.a initial public offering.b seasoned equity offering.c private-sector offering; amounts are from the Securities Data Corporation file.d group offering of multiple companies that trade separately after the IPO.

376 Journal of Economic Literature, Vol. XXXIX (June 2001)

8. Privatization’s Impact onInternational Corporate Governance

Practices

It would be an understatement to as-sert that interest in corporate gover-nance issues has been growing recentlyamong policy makers and academiceconomists. A nation’s corporate gover-nance system can be defined as the setof laws, institutions, practices, andregulations that determine how limited-liability companies will be run, and inwhose interest. Evidence of the profes-sional interest in corporate governanceis not hard to find. Several countriesand multilateral agencies have recentlypublished “codes” or “principles” ofgood corporate governance practices,such as OECD (1999). In the academicarena, one of the recent growth areas incorporate finance has been the interac-tion between law and finance, high-lighted by cross-sectional studies of thedeterminants and effects of interna-tional differences in securities law andcorporate governance. Studies have ex-amined governance practices in devel-oped countries (Shleifer and Vishny1997; LaPorta, López-de-Silanes, Shleifer,and Vishny 1997, 1998, 1999, 2000b,c;Maria Maher and Thomas Andersson1999; and Dyck 2000b); transitioneconomies (Berglof and von Thadden1999; Coffee 1999; and Dyck 2000a);and individual countries such as Russia(Black, Kraakman, and Tarassova 2000)and China (Xu and Wang 1997; and Lin2000). However, the study of the im-pact of law and corporate finance hasbeen expanded into studies of theireffects on macroeconomics (such ascurrency crisis), investment, innovation,and financing (Rajan 2000 discussesthese areas of study).

While a survey of this research is out-side the scope of this paper (Meggin-son 2000 is a more complete survey of

the effects of corporate governance),we need to mention several findings be-cause they impact the interpretation ofthe effects of privatization. There areseveral reasons analysis of internationalpatterns in corporate governance andsecurities laws has become increasinglyimportant. These include the large in-crease in the total value of security is-sues on global capital markets, and acomparable increase in the total valueof mergers and acquisitions world-wide.36 Until recently, relying on secu-rities markets for corporate financingand resorting to (often hostile) publictakeovers to effect changes in control ofcorporate assets were American prac-tices, but both trends have now “goneglobal.” In particular, the adoption ofthe euro in January 1999 has been ac-companied by the value of Europeanmergers and acquisitions roughly dou-bling to $1.22 trillion in 1999 versus1998 (itself a record year). Another rea-son for the interest in corporate gover-nance today is the important role thatpoor governance practices are per-ceived to have played in the East Asianeconomic contraction that began in July1997 (Claessens, Djankov, S. Fan, andLarry Lang 2000; and Simon Johnson,Peter Boone, Alasdair Breach, and EricFriedman 2000).

Finally, academic research by LaPorta, López-de-Silanes, and Shleifer(1999), and La Porta, López-de-Silanes,Shleifer, and Vishny (1997, 1998, 1999,2000a,c) presents evidence that corpo-rate governance generally, and corporatelegal systems specifically, significantlyinfluence capital market size, owner-ship structure, and efficiency. Most im-portantly, La Porta et al. argue in theirarticles that there are differences

36 The data are taken from the Investment Deal-ers Digest. Each January, IDD details the prioryear’s total worldwide security issuance and merg-ers and acquisitions volume.

Megginson and Netter: From State to Market 377

between countries in the degree towhich the legal system protects inves-tors, which in turn affects the develop-ment and operation of external capitalmarkets. It appears that capital marketshave developed better in countrieswhere the legal system had a commonlaw origin than in countries with a civillaw basis. Rajan (2000), however, sug-gests that some other factor that iscorrelated with the origin of the legalsystem likely explains the above findings.In any event, the framework and opera-tion of a country’s legal system im-pacts the operation of financial mar-kets and corporate governance in thatcountry.

Similarly, the structure and operationof a country’s legal system will affectthe impact of privatization. Privatiza-tion is a major change in the gover-nance structure of a firm. Thus, howwell the legal system protects investorsis presumably a determinant of the suc-cess of privatization in improving firmperformance (Sachs, Zinnes, and Eilat2000; and Djankov and Murrell 2000a,bpresent evidence this is the case in tran-sition economies). Further, privatiza-tion usually accompanies changes in acountry’s legal system. For example, in-dustrialized-country governments thatimplement large-scale SIP programsoften need to significantly change theircorporate governance systems, whilegovernments from the transition econo-mies of China and central and easternEurope must create such a system al-most from scratch. As we mentionedabove, this has an implication for mostof the studies we discuss in this paper.Since the privatizations are occurringat the same time as other majorchanges, including the legal system, itis impossible to completely isolate theimpact of privatization on firm opera-tions from the other changes affectingthe firm.

In addition, privatization impacts thepatterns of the changes in the legal sys-tem in many countries. One of the dis-tinctive aspects of SIP programs is thetendency of governments to sell sharesto large numbers of citizens, often onemillion or more. Democratic govern-ments are usually acutely aware of thepolitical fallout that could result if smallinvestors suffer losses on their SIP in-vestments because of inadequate share-holder protection or insider dealings.Thus, at the same time they launch thefirst large SIPs, most governments es-tablish (or augment) a regulatory bodysimilar to the U.S. Securities and Ex-change Commission. Since utilitiescomprise many of the important privati-zations and since many utilities arenatural monopolies, most privatizinggovernments establish regulatory bodiesfor these firms as well. In addition, na-tional stock exchanges are often illiquidand nontransparent at the beginning oflarge SIP programs. Governments mustestablish the listing and other regula-tions that will assure potential investorsthat the market is a reputable place inwhich to invest and trade.

There is some literature that exam-ines the actual corporate governanceprovisions of privatized firms. In theirstudy of SIPs, Jones, Megginson, Nash,and Netter (1999) find that govern-ments tend to retain some sort of deci-sive voting rights in privatized firmseven after a majority of the incomerights have been sold. In many coun-tries, the government retains a goldenshare (90 percent of U.K. SIPs havesuch a feature). This special share heldby the government enables it to vetomergers, liquidations, asset sales, andother major corporate events. An alter-native method of retaining ultimatecontrol is for the government to insertsome control restrictions directly intothe SIP’s charter.

378 Journal of Economic Literature, Vol. XXXIX (June 2001)

8.1 Individual Share Ownershipin Privatized Companies

Boutchkova and Megginson (2000)study the evolution of share ownershipin large SIPs. They look at how manyindividual stockholders are created in asample of large privatization share of-ferings, as well as how these highlyatomistic ownership structures evolveover time. They compare the numbersof stockholders in the privatized firmsin the 1999 Business Week “Global1000” and “Top 200 Emerging Markets”lists to capitalization-matched privatesector firms from the same markets, ob-taining useable data for 97 of the 153privatized companies and for 99 of thematching privately owned firms. Formost of the cases with data available forboth the privatized and the matchingfirm, the privatized company has alarger number of shareholders. Thisresult holds despite the fact that inmost cases governments retain sizable

stakes in these firms, thus reducingtheir effective total capitalization sincethese stakes have not yet been sold toprivate investors. Boutchkova and Meg-ginson conclude that the number ofshareholders in the privatized companiesis significantly higher than the numberof shareholders in the matchingprivate-sector (non-privatized) samplecompanies.

Boutchkova and Megginson (2000)also examine how the total number ofshareholders in a company evolves dur-ing the years subsequent to an SIP.They demonstrate that the extremelylarge numbers of shareholders createdby many SIPs are not a stable pattern ofcorporate ownership. Figure 3 showsthe dynamics in share ownership inprivatized firms. For SIPs with lessthan 100,000 initial investors, the num-ber of shareholders increases steadilyfrom one year to four years after the pri-vatization. However, for the 39 SIPsthat initially have more than 100,000

1.3

1.2

1.1

1.0

0.9

0.8

0.7

0.6

Figure 3. Changes in the Number of Shareholders in Privatized Firms over Years +1 to +6

< 100,000

Year 0 Year +1 Year +2 Year +3 Year +4 Year +5 Year +6

> 500,000

All

> 250,000

> 100,000

Notes: This figure represents the dynamics of share ownership of a sample of privatized firms, where the number ofshareholders in Year 0 is normalized to 1 and in subsequent years shows the change with respect to Year 0. The companieswith less than 100,000 initial shareholders exhibit increasing numbers of shareholders, and the companies with more than100,000, more than 250,000 and more than 500,000 initial shareholders exhibit strong declines that pull the whole sampleto a significant decrease in the number of shareholders over the whole period.

Megginson and Netter: From State to Market 379

shareholders, the total number of share-holders declines steadily. In the largestprivatizations (those with 500,000 ormore initial investors) the total num-ber of shareholders declines by 33percent within five years of the shareoffering.

The implications of this finding forgovernment efforts to develop an effec-tive corporate governance system or eq-uity culture are unclear. Many newstockholders do not retain the sharesthey purchase. Other evidence suggeststhat retail investors in privatizationsgenerally own only that one stock,hardly indicative of a class of well-diversified stockholders. On the otherhand, since the long-run returns to in-vestors in SIPs are generally positive,the first experience of these newretail investors in stock market tradingis a positive one. Furthermore, thefact that governments are able to enticelarge numbers of investors to returnfor subsequent share offerings sug-gests that these programs are indeedcreating (at least minimally) effectivegovernance systems and stock marketscapable of absorbing large new stockissues.

9. The “Lessons” of PrivatizationResearch

9.1 Some Thoughts on the CurrentLiterature

Our reading of the extant literatureon privatization suggests the followingconclusions:

1. The privatization programs of the lasttwenty years have significantly re-duced the role of state-owned enter-prises in the economic life of mostcountries. Most of this reduction hashappened in developing countriesonly during the 1990s. The SOEshare of “global GDP” has declined

from more than 10 percent in 1979 toless than 6 percent today.37

2. Research now supports the proposi-tion that privately owned firms aremore efficient and more profitablethan otherwise-comparable state-owned firms. There is limited empiri-cal evidence, especially from China,that suggests that non-privatizing re-form measures, such as price deregu-lation, market liberalization, andincreased use of incentives, can im-prove the efficiency of SOEs, but italso seems likely that these reformswould be even more effective ifcoupled with privatization.

3. Governments use three basic tech-niques to privatize their SOEs: shareissue privatizations (SIPs), asset sales,and voucher or mass privatizations.We are beginning to understand thedeterminants of the method selectedin specific circumstances. However,there is great variation within all thetechniques, because privatization is acomplex process involving a host ofpolitical and economic factors.Voucher privatizations are the leasteconomically productive divestmenttechnique, but those governmentsthat use it generally have few otherrealistic options.

4. Governments attempt to craft the of-fering terms of SIPs to balance com-peting economic, political, and finan-cial objectives. Most governmentsunderprice share offerings (particu-larly initial offerings) and then usetargeted share allocations to favor do-mestic over foreign investors. SOEemployees are particularly favored,receiving preferential allocations in91 percent of offers. Governments

37 These figures are based on the study findingsdiscussed in section 2, and on the observation thatOECD countries represent about three-quartersof world GDP and developing countries accountfor the remaining 25 percent.

380 Journal of Economic Literature, Vol. XXXIX (June 2001)

frequently retain golden shares thatgive them veto power over certaincontrol changes, and also insert vari-ous other control restrictions intothe corporate charters of privatizedfirms.

5. We know that privatization “works,”in the sense that divested firms al-most always become more efficient,more profitable, and financiallyhealthier, and increase their capitalinvestment spending. These resultshold for both transition and non-transition economies, though the re-sults vary more in the transitioneconomies. The question of whetherprivatization generally costs at leastsome SOE workers their jobs is stillunresolved. The answer is ultimatelybased on whether sales increasefaster than productivity in privatizedfirms. Most studies find that employ-ment in privatized firms usually doesfall, though three large-sample stud-ies document employment increases.What is clear is that wheneveremployment is cut, there is almostinvariably a large compensatingperformance improvement. Severalstudies also highlight the need tobring new entrepreneurial manage-ment into privatized firms to maxi-mize performance improvements.However, there is little empirical evi-dence on how privatization affectsconsumers.

6. Investors who purchase initial SIPshares at the offering price and thensell those shares at the first post-issuetrading price earn significantly posi-tive excess (market-adjusted) returns.Additionally, there is now convincingevidence that initial returns on priva-tization IPOs are significantly higherthan the initial returns earned onprivate-sector IPOs. Investors whopurchase privatization IPO shares attheir first post-offer trading price,

and then retain those shares for one-,three-, or five-year holding periods,also earn significantly positive netreturns.

7. Though it is difficult to pinpoint cau-sality, it appears that countries thathave launched large-scale SIP pro-grams have experienced rapid growthin their national stock market capi-talization and trading volume. Coun-tries (other than the United States)that have either not launched majorprivatization programs or have em-phasized asset sales and vouchersover public share offerings appear tolag behind in market development.Privatized firms are one of the two orthree most valuable companies inmost non-U.S. markets, and the tenlargest (and thirty of the 35 largest)share issues in financial history haveall been privatizations.

8. Emerging (largely anecdotal) evi-dence suggests that adopting a large-scale SIP program is often a majorspur to modernizing a nation’s corpo-rate governance system. Transitioneconomies that launch privatizationprograms must create such systemslargely from scratch, and the recordof success here is decidedly mixed.Many governments try to develop anequity culture among their citizenrythrough SIP programs, also withmixed results. Share ownership hasdramatically increased in most non-transition countries over the past fif-teen years, but the share ownershippatterns that are created when SIPsare sold to large numbers of in-vestors (often one million or more)are not stable. However, it seemsclear that privatization programs leadto significant improvements in securi-ties market regulation, informationdisclosure rules, and other requiredcomponents of modern financialsystems.

Megginson and Netter: From State to Market 381

9.2 Avenues for Further Research

While much has indeed been learnedabout the effectiveness of privatizationas a political and economic policy, thereare several important areas that needfurther research. We believe that, inparticular, there are three aspects ofprivatization that need to be understoodmuch better for public policy reasons.First, researchers need to more closelyexamine the sequencing and staging ofprivatization, and conclusively docu-ment whether reforms other than gov-ernment divestiture can effectivelyserve as a substitute (or precursor) forprivatization. Responsible policy mak-ers are understandably reluctant to “bettheir economies” on a rapid, and essen-tially irreversible, privatization programwithout some assurance that all neces-sary prerequisite policies have been putinto place. Until these policies are iden-tified, and the interactions betweenvarious policy options are established,launching large-scale privatizationprograms will remain a leap of faith.

The second vital area of research is toconclusively document the labor eco-nomics of privatization programs. Domost such programs actually cost SOEworkers jobs? Are there gender-specificimpacts relating to the total commer-cialization of state-owned enterprises,as might happen if privatization causedSOEs to shut down child care or othersocial services? Are worker training/retraining programs effective methodsof dealing with worker redundancies, orshould governments emphasize lump-sum severance packages when lay-offsare required? Do privatization pro-grams create more jobs economy-widethan they destroy? These questions arenot only vitally important to policy mak-ers, they are inherently interesting intheir own right.

Finally, what role can privatization play

in equipping companies and countriesto meet the challenges posed by majoreconomic forces such as globalizationand the rapid growth of information-based business? Technological break-throughs have transformed the globaltelecommunications industry during thepast decade, and privatized telecomcompanies have been at the forefront ofthis revolution. Indeed, it is unlikelythat this most dynamic of industrieswould have been able to grow nearly asrapidly under the former state owner-ship model. But how important will pri-vatization be for the global oil and gasindustry’s development in the future,and for the energy-based utilities thatare now being impacted by technologi-cal and regulatory changes similar tothose that hit telecommunications dur-ing the 1990s? How can developingcountries structure privatization pro-grams to most effectively attract foreigndirect investment from multinationalcompanies? How will privatization im-pact the worldwide shift from commer-cial bank-based systems of corporate fi-nance to capital market-based finance?All of these questions can, and should,be answered using the tools of eco-nomic analysis, and it is hard to imaginean area of research more intrinsically in-teresting to economists than analyzingthe optimal role of government in thebusiness of nations.

REFERENCES

Aggarwal, Reena; Ricardo Leal, and LeonardoHernandez. 1993. “The Aftermarket Perfor-mance of Initial Public Offerings in LatinAmerica, Finan. Manage. 22, pp. 43–53.

Alexandrowicz, Melinda Roth. 1994. “Mass Privati-zation Programs,” FPD Note No. 4, Washington,DC: World Bank.

Alchain, Armen. 1977. “Some Economics of Prop-erty Rights,” in Economic Forces at Work. Ar-men Alchain, ed. Indianapolis: Liberty Press,pp. 127–149.

Allen, Franklin and Douglas Gale. 1999. “Corpo-rate Governance and Competition,” working pa-per, Wharton School, U. Pennsylvania.

Aussenegg, Wolfgang. 2000. “Privatization versus

382 Journal of Economic Literature, Vol. XXXIX (June 2001)

Private Sector Initial Public Offerings in Po-land,” working paper, Vienna U. Technology.

Bai, Chong-en; David D. Li and Yijaiang Wang.1997. “Enterprise Productivity and Efficiency:When Is Up Really Down?” J. Comp. Econ. 24,pp. 265–80.

Baldwin, Carliss Y. and Sugato Bhattacharya.1991. “Choosing the Method of Sale: A ClinicalStudy of Conrail,” J. Finan. Econ. 30, pp. 69–98.

Barber, Brad and John Lyon. 1996. “Detecting Ab-normal Operating Performance: The EmpiricalPower and Specification of Test Statistics,” J.Finan. Econ. 41, pp. 359–99.

———. 1997. “Detecting Long-Run AbnormalStock Returns: The Empirical Power and Speci-fication of Test Statistics,” J. Finan. Econ. 43,pp. 341–72.

Barberis, Nicholas; Maxim Boycko, AndreiShleifer and Natalia Tsukanova. 1996. “HowDoes Privatization Work? Evidence from theRussian Shops,” J. Polit. Econ. 104, pp. 764–90.

Beck, Thorsten; Ross Levine, and Norman Loayza.2000. “Finance and the Sources of Growth,” J.Finan. Econ. 58, pp. 261–300.

Beesley, M. E. and S. C. Littlefield. 1989. “TheRegulation of Privatized Monopolies in theUnited Kingdom,” Rand J. Econ. 20, pp. 454–73.

Bekaert, Geert and Campbell Harvey. 2000. “For-eign Speculators and Emerging Equity Mar-kets,” J. Fin. 55, pp. 565–613.

Bennell, Paul. 1997. “Privatization in Sub-SaharanAfrica: Progress and Prospects during the1990s,” World Develop. 25, pp. 1785–803.

Berg, Andrew; Eduardo Borensztein, Ratna Sahay,and Jeromin Zettelmeyer. 1999. “The Evolutionof Output in Transition Economies: Explain-ing the Differences,” IMF working paperWP/99/73.

Berglof, Eric and Gérard Roland. 1998. “SoftBudget Constraints and Banking in TransitionEconomies,” J. Comp. Econ. 26, pp. 18–40.

Berglof, Eric and Ernst-Ludwig von Thadden.1999. “The Changing Corporate GovernanceParadigm: Implications for Transitional and De-veloping Countries,” SITE working paper,Stockholm School of Economics.

Biais, Bruno and Enrico Perotti. 2000. “Machiavel-lian Privatization,” forthcoming, Amer. Econ.Rev.

Bishop, Matthew R. and John A. Kay. 1989. “Pri-vatization in the United Kingdom: Lessons fromExperience,” World Develop. 17, pp. 643–57.

Black, Bernard; Reinier Kraakman, and AnnaTarassova. 2000. “Russian Privatization andCorporate Governance: What Went Wrong?”Stanford Law Rev. 52, forthcoming.

Blanchard, Olivier and Philippe Aghion. 1996. “OnInsider Privatization,” Europ. Econ. Rev. 40,pp. 759–66.

Boardman, Anthony E. and Claude Laurin. 2000.“Factors Affecting the Stock Price Performanceof Share Issue Privatizations,” Applied Econ.forthcoming.

Boardman, Anthony E.; Claude Laurin and AidanVining. 2000. “Privatization in Canada: Operat-ing, Financial and Stock Price Performancewith International Comparisons,” working pa-per, U. British Columbia, Vancouver.

Boardman, Anthony and Aidan R. Vining. 1989.“Ownership and Performance in CompetitiveEnvironments: A Comparison of the Perfor-mance of Private, Mixed, and State-Owned En-terprises,” J. Law Econ. 32:1, pp. 1–33.

Boles de Boer, David and Lewis Evans. 1996.“The Economic Efficiency of Telecommunica-tions in a Deregulated Market: The Case ofNew Zealand,” Econ. Record 72, pp. 24–35.

Bornstein, Morris. 1994. “Russia’s Mass Privatiza-tion Program,” Communist Economies Econ.Transformation, 6:4, pp. 419–57.

———. 1999. “Framework Issues in the Privatiza-tion Strategies of the Czech Republic, Hungaryand Poland,” Post-Communist Econ. 11:1, pp.47–77.

Bortolotti, Bernardo; Marcella Fantini, and CarloScarpa. 2000. “Why Do Governments Sell Priva-tised Companies Abroad?” working paper, Fon-dazione ENI-Enrico Mattei (FEEM), Milan.

Bortolotti, Bernardo; Marcella Fantini, andDomenico Siniscalco. 1999a. “Privatisation:Politics, Institutions, and Financial Markets,”working paper, FEEM, Milan.

———. 1999b. “Regulation and Privatisation: TheCase of Electricity,” working paper, FEEM,Milan.

Boubakri, Narjess and Jean-Claude Cosset. 1998.“The Financial and Operating Performance ofNewly-Privatized Firms: Evidence from Devel-oping Countries,” J. Fin. 53, pp. 1081–110.

———. 1999. “Does Privatization Meet the Expec-tations? Evidence from African Countries,”working paper, Ecole des HEC, Montreal.

———. 2000. “The Aftermarket Performance ofPrivatization Offerings in Developing Coun-tries,” working paper, Ecole des HEC, Mon-treal.

Boutchkova, Maria K. and William L. Megginson.2000. “The Impact of Privatization on CapitalMarket Development and Individual ShareOwnership,” Finan. Manage. 29:1, pp. 67–77.

Boycko, Maxim; Andrei Shleifer, and Robert W.Vishny. 1994. “Voucher Privatization,” J. Finan.Econ. 35, pp. 249–66.

———. 1996a. “A Theory of Privatisation,” Econ.J. 106, pp. 309–19.

———. 1996b. “Second-Best Economic Policy fora Divided Government,” Europ. Econ. Rev. 40,pp. 767–74.

Boylaud, Olivier and Giuseppe Nicoletti. 2000.“Regulation, Market Structure and Perfor-mance in Telecommunications,” OECD Econ.Dept. working paper 237, Paris.

Brada, Josef C. 1996. “Privatization Is Transi-tion—Or Is It?” J. Econ. Perspect. 10, pp. 67–86.

Brav, Alon and Paul A. Gompers. 1997. “Myth orReality? The Long-Run Underperformance of

Megginson and Netter: From State to Market 383

Initial Public Offerings: Evidence from Ven-ture- and Non-Venture Capital-Backed Compa-nies,” J. Fin. 52, pp. 1791–821.

Brav, Alon; Christopher Geczy, and Paul Gom-pers. 2000. “Is the Abnormal Return FollowingEquity Issuances Abnormal?” J. Finan. Econ.56, pp. 209–49.

Brickley, James A.; Clifford W. Smith, and JeroldL. Zimmerman. 2001. Managerial Economics andOrganizational Architecture. 2nd ed. Chicago:Irwin.

Brickley, James A. and R. Lawrence Van Horn. 2000.“Incentives in Nonprofit Organizations: Evidencefrom Hospitals,” working paper, U. Rochester.

Briggs, A. 1961. “The Welfare State in HistoricalPerspective,” Archives Europ. Sociologie 2:2,pp. 221–59.

Bulow, Jeremy and Paul Klemperer. 1996. “Auc-tions versus Negotiations,” Amer. Econ. Rev.86, pp. 180–94.

Canina, Linda; Roni Michaely, Richard Thaler,and Kent Womack. 1998. “Caveat Compounder:A Warning about Using the Daily CRSP Equal-Weighted Index to Compute Long-Run ExcessReturns,” J. Fin. 53, pp. 403–16.

Cao, Yuanzheng; Yingyi Qian, and Barry R. Wein-gast. 1999. “From Federalism, Chinese Style toPrivatization, Chinese Style,” Econ.Transition7:1, pp. 103–31.

Carter, Richard B.; Frederick H. Dark, and AjaiK. Singh. 1998. “Underwriter Reputation, Ini-tial Returns, and the Long-Run Performance ofIPO Stocks,” J. Fin. 53, pp. 285–311.

Choi, Seung-Doo and Sang-Koo Nam. 2000. “TheShort-Run Performance of IPOs of Privatelyand Publicly Owned Firms: International Evi-dence,” Multinat. Finance J. forthcoming.

Choi, Seung-Doo; Sang-Koo Nam, and Gui-YoulRyu. 2000. “Do Privatization IPOs Outperformthe Market? International Evidence,” workingpaper, Korea U., Seoul.

Claessens, Stijn. 1997. “Corporate Governanceand Equity Prices: Evidence from the Czechand Slovak Republics,” J. Fin. 52, pp. 1641–58.

Claessens, Stijn and Simeon Djankov. 1999a. “En-terprise Performance and Management Turn-over in the Czech Republic,” Europ. Econ. Rev.43, pp. 1115–24.

———. 1999b. “Ownership Concentration andCorporate Performance in the Czech Repub-lic,” J. Comp. Econ. 27, pp. 498–513.

Claessens, Stijn; Simeon Djankov, S. Fan, andLarry Lang. 2000. “Expropriation of MinorityShareholders: Evidence from East Asia,” WorldBank Policy Research Paper 2088.

Claessens, Stijn; Simeon Djankov, and GerhardPohl. 1997. “Ownership and Corporate Gover-nance: Evidence from the Czech Republic,”World Bank Policy Research Paper 1737.

Coase, Ronald. 1960. “The Problem of SocialCost,” J. Law Econ. 1, pp. 1–44.

Coffee, Jack C. Jr. 1999. “Privatization and Corpo-rate Governance: The Lessons From SecuritiesMarket Failure,” J. Corporat. Law 25, pp. 1–39.

Cornelli, Francesca and David D. Li. 1997. “LargeShareholders, Private Benefits of Control, andOptimal Schemes of Privatization,” Rand J.Econ. 28, pp. 585–604.

Cragg, Michael I. and I. J. Alexander Dyck. 1999a.“Management Control and Privatization in theUnited Kingdom,” Rand J. Econ. 30:3, pp. 475–497.

———. 1999b. “Privatization, Compensation andManagement Incentives: Evidence From theUnited Kingdom,” working paper, HarvardBusiness School.

Davidson, Richard. 1998. “Market Analysis: Un-derperformance Over?” Privatisation Interna-tional Yearbook. London: IFR Publishing.

Davis, Jeffry; Rolando Ossowski, Thomas Richard-son, and Steven Barnett. 2000. “Fiscal andMacroeconomic Aspects of Privatization,” IMFOccasional Paper 194.

Demirgüç-Kunt, Asli and Vojislav Maksimovic.1998. “Law, Finance, and Firm Growth,” J. Fin.53, pp. 2107–39.

Dewenter, Kathryn and Paul H. Malatesta. 1997.“Public Offerings of State-Owned and PrivatelyOwned Enterprises: An International Compari-son,” J. Fin. 52, pp. 1659–79.

———. 2001. “State-Owned and Privately-OwnedFirms: An Empirical Analysis of Profitability,Leverage, and Labour Intensity,” Amer. Econ.Rev. 91, pp. 320–34.

Djankov, Simeon. 1999a. “Ownership Structureand Enterprise Restructuring in Six Newly In-dependent States,” Comp. Econ. Stud. 41:1, pp.75–95.

———. 1999b. “The Restructuring of Insider-Dominated Firms: A Comparative Analysis,”Econ. Transition 7:2, pp. 467–79.

Djankov, Simeon and Peter Murrell, 2000a. “TheDeterminants of Enterprise Restructuring inTransition: An Assessment of the Evidence,”working paper, U. Maryland.

———. 2000b. “Enterprise Restructuring in Tran-sition: A Quantitative Survey,” working paper,U. Maryland.

Drum, Bernard. 1994. “Mass Privatization inUkraine,” FPD Note No. 8, World Bank.

D’Souza, Juliet and William L. Megginson. 1999.“The Financial and Operating Performance ofNewly Privatized Firms in the 1990s,” J. Fin.54, pp. 1397–438.

———. 2000. “Sources of Performance Improve-ment in Privatized Firms: A Clinical Study ofthe Global Telecommunications Industry,”working paper, U. Oklahoma.

D’Souza, Juliet; Robert Nash, and William L.Megginson. 2000. “Determinants of Perfor-mance Improvement in Newly-PrivatizedFirms: Does Restructuring and CorporateGovernance Matter?” working paper, U. Okla-homa.

Dyck, I. J. Alexander. 1997. “Privatization in East-ern Germany: Management Selection and Eco-nomic Transition,” Amer. Econ. Rev. 87, pp.565–97.

384 Journal of Economic Literature, Vol. XXXIX (June 2001)

———. 2000a. “Ownership Structure, Legal Pro-tections and Corporate Governance,” workingpaper, Harvard Business School.

———. 2000b. “Privatization and CorporateGovernance: Principles, Evidence and FutureChallenges,” working paper, Harvard BusinessSchool.

Earle, John. 1998. “Post-Privatization Ownershipand Productivity in Russian Industrial Enter-prises,” SITE working paper 127, StockholmSchool of Economics.

Earle, John and Saul Estrin 1998. “Privatization,Competition and Budget Constraints: Disciplin-ing Enterprises in Russia,” SITE working paper128, Stockholm School of Economics.

Eckbo, B. Espen, Ronald Masulis and ØyvindNorli. 2000. “Seasoned Public Offerings: Reso-lution of the “New Issues Puzzle,” J. Finan.Econ. 56, pp. 251–91.

Eckel, Catherine; Doug Eckel, and Vijay Singal.1997. “Privatization and Efficiency: IndustryEffects of the Sale of British Airways,” J. Finan.Econ. 43, pp. 275–98.

Ehrlich, Isaac; Georges Gallais-Hamonno,Zhiqiang Liu, and Randall Lutter. 1994. “Pro-ductivity Growth and Firm Ownership: An Em-pirical Investigation,” J. Polit. Econ. 102, pp.1006–38.

Fama, Eugene F. 1998. “Market Efficiency, Long-Term Returns, and Behavioral Finance,” J. Fi-nan. Econ. 49, pp. 283–306.

Foerster, Stephen R. and G. Andrew Karolyi.2000. “The Long-Run Performance of GlobalEquity Offerings,” J. Finan. Quant. Anal. forth-coming.

Frydman, Roman; Cheryl W. Gray, Marek Hessel,and Andrzej Rapaczynski. 1999. “When DoesPrivatization Work? The Impact of PrivateOwnership on Corporate Performance in Tran-sition Economies,” Quart. J. Econ. 114:4, pp.1153–91.

———. 2000. “The Limits of Discipline: Owner-ship and Hard Budget Constraints in the Tran-sition Economies,” C. V. Starr Center forApplied Econ. working paper, NYU.

Frydman, Roman; Marek Hessel, and Andrzej Ra-paczynski. 2000. “Why Ownership Matters: En-trepreneurship and the Restructuring of Enter-prises in Central Europe,” C. V. Starr Centerfor Applied Econ. working paper, NYU.

Frydman, Roman; Katharina Pistor, and AndrzejRapaczynski. 1996. “Exit and Voice After MassPrivatization: The Case of Russia,” Europ.Econ. Rev. 40, pp. 581–588.

Galal, Ahmed; Leroy Jones, Pankaj Tandon, andIngo Vogelsang. 1994. Welfare Consequences ofSelling Public Enterprises. Oxford: Oxford U.Press.

Gibbon, Henry. 1997. “A Seller’s Manual: Guide-lines for Selling State-Owned Enterprises,” Pri-vatisation Yearbook, Privatisation Int. pp. 16–26.

———. 1998. “Worldwide Economic Orthodoxy,”Privatisation Int. 123, pp. 4–5.

———. 2000. “Editor’s Letter,” PrivatisationYearbook, London: Thomson Financial, p. 1.

Gottret, Pablo. 1999. “Bolivia: Capitalisation, Pen-sion Reform, and Their Impact on Capital Mar-kets,” working paper, OECD, Paris.

Gough, I. 1989. “The Welfare State,” in The NewPalgrave Social Economics. NY: W. W. Norton,pp. 276–81.

Groves, Theodore; Yongmiao Hong, John McMil-lan, and Barry Naughton. 1994. “Autonomy andIncentives in Chinese State Enterprises,”Quart. J. Econ. 109:1, pp. 183–209.

———. 1995. “China’s Evolving Managerial LaborMarket,” J. Polit. Econ. 103:4, pp. 873–92.

Hansmann, Henry and Reinier Kraakman. 2000.“The End of History for Corporate Law,” work-ing paper, NYU and Harvard Law School.

Harper, Joel T. 2000. “The Performance of Priva-tized Firms in the Czech Republic,” workingpaper, Florida Atlantic U., Boca Raton.

Havrylyshyn, Oleh and Donald McGettigan. 2000.“Privatisation in Transition Countries,” Post So-viet Affairs 16, pp. 257–86.

Hayek, Friedrich von. 1994. The Road to Serfdom.Chicago: U. Chicago Press.

Henry, Peter. 2000a. “Do Stock Market Liberali-zations Cause Investment Booms?” J. Finan.Econ. 58, pp. 301–34.

———. 2000b. “Stock Market Liberalization, Eco-nomic Reform, and Emerging Market EquityPrices,” J. Fin. 55, pp. 529–64

Hersch, Philip; David Kemme, and Jeffry Netter.1997. “Access to Bank Loans in a TransitionEconomy: The Case of Hungary,” J. Comp.Econ. 24, pp. 79–89.

Hingorani, Archana; Kenneth Lehn, and AnilMakhija. 1997. “Investor Behavior in Mass Pri-vatization: The Case of the Czech VoucherScheme,” J. Finan. Econ. 44, pp. 349–96.

Huang, Qi and Richard M. Levich. 1998. “Under-pricing of New Equity Offerings by PrivatizedFirms: An International Test,” working paper,NYU.

Ibbotson, Roger; Jody Sindelar, and Jay Ritter.1994. “The Market’s Problem with the Pricingof Initial Public Offerings,” J. App. Corp. Fin.7, pp. 66–74.

Jelic, Ranko and Richard Briston. 2000a. “Pri-vatisation Initial Public Offerings: The PolishExperience,” working paper, U. Birmingham,U.K.

———. 2000b. “Hungarian Privatisation Strategyand Financial Performance of Privatised Com-panies,” J. Bus. Fin. Acc. forthcoming.

Jenkinson, Timothy and Colin Mayer. 1988. “ThePrivatisation Process in France and the U.K.”Europ. Econ. Rev. 32, pp. 482–90.

Johnson, Simon; Peter Boone, Alasdair Breach,and Eric Friedman. 2000. “Corporate Gover-nance in the Asian Financial Crisis,” J. Finan.Econ. 58, pp. 141–86.

Jones, Leroy P. 1991, “Performance Evaluation forPublic Enterprises,” World Bank DiscussionPaper 122, Washington DC.

Megginson and Netter: From State to Market 385

Jones, Steven L.; William L. Megginson, RobertC. Nash, and Jeffry M. Netter. 1999. “Share Is-sue Privatizations as Financial Means to Politi-cal and Economic Ends,” J. Finan. Econ. 53,pp. 217–53.

Karpoff, Jonathan. 2001. “Public versus PrivateInitiative in Arctic Exploration: The Effects ofIncentives and Organizational Form,” J. Polit.Econ. 109, pp. 38–78.

Katz, Barbara G. and Joel Owen. 1993. “Privatiza-tion: Choosing the Optimal Time Path,” J. Com-par. Econ. 17, pp. 715–36.

———. 1995. “Designing the Optimal Privatiza-tion Plan for Restructuring Firms and Indus-tries in Transition,” J. Compar. Econ. 21, pp. 1–28.

———. 1997. “Optimal Voucher PrivatizationFund Bids when Bidding Affects Firm Perfor-mance,” J. Compar. Econ. 24, pp. 25–43.

Kay, J. A. and D. J. Thompson. 1986. “Privatisa-tion: A Policy in Search of a Rationale,” Econ. J.96, pp. 18–32.

Kikeri, Sunita; John Nellis, and Mary Shirley.1992. Privatization: The Lessons of Experience.Washington, DC: World Bank.

Kole, Stacey R. and J. Harold Mulherin. 1997.“The Government as a Shareholder: A Casefrom the United States,” J. Law Econ. 40, pp.1–22.

Kornai, János. 1988. “Individual Freedom and Re-form of the Socialist Economy,” Europ. Econ.Rev. 32, pp. 233–67.

———. 1993. “The Evolution of Financial Disci-pline under the Postsocialist System,” Kyklos46:3, pp. 315–36.

———. 2000. “Ten Years after ‘The Road to aFree Economy’: The Author’s Self Evaluation,”working paper, Harvard U.

Kothari, S. P. and Jerold B. Warner. 1997. “Mea-suring Long-Horizon Security Price Perfor-mance,” J. Finan. Econ. 43, pp. 301–40.

La Porta, Rafael and Florencio López-de-Silanes.1999. “Benefits of Privatization—Evidence FromMexico,” Quart. J. Econ. 114:4, pp. 1193–242.

La Porta, Rafael; Florencio López-de-Silanes, andAndrei Shleifer. 2000a. “Government Owner-ship of Banks,” NBER Working Paper 7620,Cambridge, MA.

———. 2000b. “Investor Protection and CorporateGovernance,” J. Finan. Econ. 58, pp. 3–27.

La Porta, Rafael; Florencio López-de-Silanes, An-drei Shleifer, and Robert Vishny. 1997. “LegalDeterminants of External Finance,” J. Fin. 52,pp. 1131–50.

———. 1998. “Law and Finance,” J. Polit. Econ.106, 1113–1150.

———. 1999. “The Quality of Government,” J.Law Econ. Org. 15:1, pp. 222–79.

———. 2000. “Investor Protection and CorporateValuation,” NBER Working Paper 7403, Cam-bridge, MA.

Laffont, Jean-Jacques and Jean Tirole. 1993. ATheory of Incentives in Procurement and Regu-lation. Cambridge, MA: MIT Press.

Lau, Lawrence J. Yingyi Qian, and Gerard Roland.2000. “Reform Without Losers: An Interpreta-tion of China’s Dual-Track Approach to Transi-tion,” J. Polit. Econ. 108, pp. 120–143.

Laurin, Claude and Yves Bozec. 2000. “Privatiza-tion and Productivity Improvement: The Caseof Canadian National (CN),” working paper,Ecoles de HEC, Montreal.

Levine, Ross. 1997. “Financial Development andEconomic Growth: Views and Agenda,” J. Econ.Lit. 35, pp. 688–726.

Levine, Ross and Sara Zervos. 1998. “Stock Mar-kets, Banks, and Economic Growth,” Amer.Econ. Rev. 88, pp. 537–558.

Levis, Mario. 1993. “The Long-Run Performanceof Initial Public Offerings: The U.K. Experi-ence 1980–88,” Finan. Manag. 22, pp. 28–41.

Li, Wei. 1997. “The Impact of Economic Reformon the Performance of Chinese State Enter-prises, 1980–1989,” J. Polit. Econ. 105, pp.1080–106.

Lin, Cyril. 2000. “Corporate Governance of State-Owned Enterprises in China,” working paper,Asian Development Bank, Manila.

Lin, Justin Yifu; Fang Cai, and Zhou Li. 1998.“Competition, Policy Burdens, and State-Owned Enterprise Reform,” Amer. Econ. Rev.88, pp. 422–27.

Lizal, Lubomir; Miroslav Singer, and Jan Svejnar,2000, “Enterprise Break-ups and PerformanceDuring the Transition from Plan to Market,”Rev. Econ. Statist. forthcoming.

Ljungqvist, Alexander P.; Tim Jenkinson, and Wil-liam J. Wilhelm, Jr. 2000. “Has the Introductionof Bookbuilding Increased the Efficiency of In-ternational IPOs?” working paper, Stern Schoolof Business, NYU.

López-de-Silanes, Florencio. 1997. “Determinantsof Privatization Prices,” Quart. J. Econ. 112, pp.965–1025.

López-de-Silanes, Florencio; Andrei Shleifer, andRobert W. Vishny. 1997. “Privatization inthe United States,” Rand J. Econ. 28, pp. 447–71.

Loughran, Tim and Jay R. Ritter. 1995. “The NewIssues Puzzle,” J. Fin. 50, pp. 23–51.

———. 1997. “The Operating Performance ofFirms Conducting Seasoned Equity Offerings,”J. Finance 52, pp. 1823–50.

Loughran, Tim; Jay Ritter, and Kristian Rydqvist.1994. “Initial Public Offerings: International In-sight,” Pacific-Basin Finan. J. 2, pp. 165–99.

Lyon, John; Brad Barber, and Chih-Ling Tsai.1999. “Improved Methods for Tests of Long-Run Abnormal Stock Returns,” J. Finance 54,pp. 165–201.

Macquieira, Carlos and Salvador Zurita. 1996.“Privatizaciones en Chile: Eficiencia y PoliticasFinancieras,” Estudios de Administracion 3:2,pp. 1–36.

Mahboobi, Ladan. 2000. “Recent PrivatisationTrends,” OECD Financial Market Trends, No.76, pp. 43–64.

Maher, Maria and Thomas Andersson. 1999,

386 Journal of Economic Literature, Vol. XXXIX (June 2001)

“Corporate Governance: Effects on Firm Per-formance and Economic Growth,” workingpaper, OECD, Paris.

Majumdar, Sumit K. 1996. “Assessing Compara-tive Efficiency of the State-Owned, Mixed, andPrivate Sectors in Indian Industry,” PublicChoice 96, pp. 1–24.

Martin, Stephen and David Parker. 1995. “Privati-zation and Economic Performance Throughoutthe UK Business Cycle,” Manag. Decis. Econ.16, pp. 225–37.

McCurry, Patrick. 2000. “Golden Shares Fail toShine,” Privatisation Int. 136, pp. 41–43.

Megginson, William. 2000. “Corporate Gover-nance in Publicly Quoted Companies,” workingpaper, OECD, Paris.

Megginson, William; Robert Nash, Jeffry Netter,and Annette Poulsen. 2000. “The Choice be-tween Private and Public Markets: Evidencefrom Privatizations,” working paper, U. Geor-gia.

Megginson, William; Robert. Nash, Jeffry Netter,and Adam Schwartz. 2000. “The Long Term Re-turn to Investors in Share Issue Privatizations,”Finan. Manag. 29, pp. 67–77.

Megginson, William; Robert Nash, and Matthiasvan Randenborgh. 1994. “The Financial andOperating Performance of Newly PrivatizedFirms: An International Empirical Analysis,” J.Finance 49, pp. 403–52.

Menyah, Kojo and Krishna Paudyal. 1996. “ShareIssue Privatisations: The UK Experience,” inEmpirical Issues in Raising Equity Capital.Mario Levis, ed. Amsterdam: Elsevier Science.

Menyah, Kojo; Krishna Paudyal, and Charles In-yangete. 1995. “Subscriber Return, Underpric-ing, and Long-Term Performance of U.K. Priva-tization Initial Public Offers,” J. Econ. Bus. 47,pp. 473–95.

Mihályi, Peter. 2000. “FDI Through Cross-BorderM&A—The Post-Communist Privatization StoryReconsidered,” working paper, UNCTAD, NewYork.

Mitchell, Mark and Erik Stafford. 2000. “Manage-rial Decisions and Long-Term Stock Price Per-formance,” J. Bus. 73, pp. 287–329.

Nellis, John. 1994. “Is Privatization Necessary?”Viewpoint Note 17, Washington DC: WorldBank.

———. 1999. “Time to Rethink Privatization inTransition Economies?” IFC Discussion Paper38, World Bank Group.

———. 2000. “Privatization in Transition Econo-mies: What Next?” working paper, World Bank.

Nellis, John and Sunita Kikeri. 1989. “Public En-terprise Reform: Privatization and the WorldBank,” World Devel. 17, pp. 659–72.

Newbery, David. 1997. “Privatization and Liberali-sation of Network Utilities,” Europ. Econ. Rev.41, pp. 357–83.

Newbery, David and Michael G. Pollitt. 1997.“The Restructuring and Privatization of Brit-ain’s CEGB—Was It Worth It?,” J. Indust.Econ. 45, pp. 269–303.

Noll, Roger G. 2000. “Telecommunications Re-form in Developing Countries,” in EconomicPolicy Reform: The Second Stage. Anne O.Kreuger, ed. Chicago: U. Chicago Press.

Olds, Kelly, 1994. “Privatizing the Church: Dises-tablishment in Connecticut and Massachusetts,”J. Polit. Econ. 102, pp. 277–97.

Organization of Economic Cooperation and Devel-opment. 1999. OECD Principles of CorporateGovernance, Paris: OECD.

Paudyal, K.; B. Saadouni and R. J. Briston. 1998.“Privatization Initial Public Offerings in Malay-sia: Initial Premium and Long-Term Perfor-mance,” Pacific Basin Fin. J. 6. pp. 427–51.

Peltzman, Sam. 1971. “Pricing in Public and Pri-vate Enterprises: Electric Utilities in theUnited States,” J. Law Econ. 14, pp. 109–47.

Perotti, Enrico. 1995. “Credible Privatization,”Amer. Econ. Rev. 85, pp. 847–59.

Perotti, Enrico and Serhat E. Guney. 1993. “Suc-cessful Privatization Plans: Enhanced Credibil-ity Through Timing and Pricing of Sales,” Fin.Manage. 22, pp. 84–98.

Perotti, Enrico and Pieter van Oijen. 2000. “Priva-tization, Political Risk and Stock Market Devel-opment in Emerging Economies,” J. Int. MoneyFin. forthcoming.

Petrazzini, Ben A. and Theodore H. Clark. 1996.“Costs and Benefits of TelecommunicationsLiberalization in Developing Countries,” work-ing paper, Hong Kong U. Science Technology.

Pinto, Brian; Merek Belka, and Stefan Krajewski.1993. “Transforming State Enterprises in Po-land: Evidence on Adjustment by Manufactur-ing Firms,” Brookings Papers Econ. Act. 0:1,pp. 213–61.

Pistor, Katharina and Andrew Spicer. 1996. “In-vestment Funds in Mass Privatization andBeyond: Evidence from the Czech Republicand Russia,” Private Sector 7, World Bank, pp.33–36.

Pohl, Gerhard; Robert E. Anderson, Stijn Claes-sens, and Simeon Djankov. 1997. “Privatizationand Restructuring in Central and EasternEurope: Evidence and Policy Options,” WorldBank Technical Paper 368.

Pollitt, Michael G. 1997. “The Impact of Liberali-zation on the Performance of the Electric Sup-ply Industry: An International Survey,” J. En-ergy Lit. 3:2, pp. 3–31.

Price Waterhouse. 1989a. Privatization: Learningthe Lessons from the U.K. Experience. London:Price Waterhouse.

———. 1989b. Privatization: The Facts. London:Price Waterhouse.

Rajan, Raghuram. 2000. NBER Program Report:Corporate Finance.” NBER Reporter, Spring2000, pp. 1–5.

Rajan, Raghuram G. and Luigi Zingales. 1998. “Fi-nancial Dependence and Growth,” Amer. Econ.Rev. 88, pp. 559–86.

Ramamurti, Ravi. 1996. “The New Frontier of Pri-vatization,” in Privatizing Monopolies: Lessonsfrom the Telecommunications and Transport

Megginson and Netter: From State to Market 387

Sectors in Latin America. Ravi Ramamurti, ed.Baltimore: Johns Hopkins U. Press, pp. 1–45.

———. 1997. “Testing the Limits of Privatization:Argentine Railroads,” World Develop. 25, pp.1973–93.

Ritter, Jay R. 1991. “The Long-Run Performanceof Initial Public Offerings,” J. Fin. 46, pp. 3–27.

Rondinelli, Dennis and Max Iacono. 1996. Policiesand Institutions for Managing Privatization. In-ternational Training Centre, ILO, Turin, Italy.

Ros, Agustin J. 1999. “Does Ownership or Compe-tition Matter? The Effects of Telecommunica-tions Reform on Network Expansion and Effi-ciency,” J. Reg. Econ. 15, pp. 65–92.

Sachs, Jeffrey; Clifford Zinnes, and Yair Eilat.2000. “The Gains from Privatization in TransitionEconomies: Is ‘Change of Ownership’ Enough?”CAER II Discussion paper 63, HIID, Harvard.

Samonis, Val. 2000. “Mergers and Acquisitions inTransition Economies: ‘Williams’ LithuaniaDeal Decomposed,” working paper, Center forEuropean Integration Studies, Bonn, Germany.

Sappington, David E.M. and Joseph E. Stiglitz.1987. “Privatization, Information and Incen-tives,” J. Pol. Anal. Manage. 6, pp. 567–82.

Schaffer, Mark E. 1998. “Do Firms in TransitionEconomies Have Soft Budget Constraints? AReconsideration of Concepts and Evidence,” J.Comp. Econ. 26, pp. 80–103.

Schmidt, Klaus. 1996. “The Costs and Benefits ofPrivatization: An Incomplete Contracts Ap-proach,” J. Law Econ. Org. 12:1, pp. 1–24.

Schmitz, James A. Jr. 1996. “The Role of PublicEnterprises: How Much Does It Differ acrossCountries?” Fed. Reserve Bank of MinneapolisQuart. Rev. 20, pp. 2–15.

Shafik, Nemat. 1995. “Making a Market: Mass Pri-vatization in the Czech and Slovak Republics,”World Develop. 23, pp. 1143–56.

———. 1996. “Selling Privatization Politically,”Columbia J. World Bus. 31, pp. 20–29.

Sheshinski, Eytan and Luis Felipe López-Calva.1999. “Privatization and its Benefits: Theoryand Evidence,” HIID Discussion Paper 698,Harvard U.

Shirley, Mary M. 1999. “Bureaucrats in Business:The Role of Privatization in State Owned En-terprise Reform,” World Develop. 27:1, pp.115–36.

Shirley, Mary and John Nellis, 1991, “Public En-terprise Reform: The Lessons of Experience,”Publication 9800, World Bank, Washington, DC.

Shirley, Mary and Patrick Walsh. 2000. “Public vs.Private Ownership: The Current State of theDebate,” working paper, World Bank, Washing-ton, DC.

Shirley, Mary M. and Lixin Colin Xu. 1998. “Infor-mation, Incentives, and Commitment: An Em-pirical Analysis of Contracts between Govern-ment and State Enterprises,” J. Law Econ. Org.14:2, pp. 358–78.

Shleifer, Andrei. 1998. “State versus Private Own-ership, J. Econ. Perspect. 12, pp. 133–50.

Shleifer, Andrei and Robert W. Vishny. 1997. “A

Survey of Corporate Governance,” J. Finance52, pp. 737–83.

Smith, Peter L. and Björn Wellenius. 1999. “Miti-gating Regulatory Risk in Telecommunica-tions,” Private Sector, World Bank, pp. 33–44.

Smith, Stephen C.; Beon-Cheol Cin, and MilanVodopivec. 1997. “Privatization Incidence, Owner-ship Forms, and Firm Performance: Evidencefrom Slovenia,” J. Comp. Econ. 25, pp. 158–79.

Sobel, Robert. 1999. The Pursuit of Wealth. NewYork: McGraw Hill.

Spiess, D. Katherine and John Affleck-Graves.1995. “Underperformance in Long-run StockReturns Following Seasoned Equity Offerings,”J. Finan. Econ. 38, pp. 243–67.

Stigliz, Joseph. 1998. “The Private Uses of PublicInterests: Incentives and Institutions,” J. Econ.Perspect. 12, pp. 3–22.

Su, Dongwei and Belton M. Fleisher. 1999. “AnEmpirical Investigation of Underpricing in Chi-nese IPOs,” Pacific Basin Fin. J. 7, pp. 173–202.

Subrahmanyam, Avanidhar and Sheridan Titman. 1999.“The Going Public Decision and the Developmentof Financial Markets,” J. Finance 54, pp. 1045–82.

Temple, Jonathan. 1999. “The New Growth Evi-dence,” J. Econ. Lit. 37, pp. 112–156.

Tian, George Lihui. 2000. “State Shareholding andCorporate Performance: A Study of a UniqueChinese Data Set,” working paper, LondonBusiness School.

Tybout, James. 2000. “Manufacturing Firms inDeveloping Countries: How Well Do They Doand Why?” J. Econ. Lit. 38, pp. 11–44.

Verbrugge, James; Wanda Owens, and WilliamMegginson. 2000. “State Ownership and the Fi-nancial Performance of Privatized Banks: AnEmpirical Analysis,” Proceedings of a Policy Re-search Workshop at the World Bank, March15–16, 1999. Dallas: Fed. Reserve Bank Dallas.

Vickers, John and George Yarrow. 1991. “Eco-nomic Perspectives on Privatization,” J. Econ.Perspect. 5, pp. 111–32.

Vining, Aidan R. and Anthony E. Boardman. 1992.“Ownership Versus Competition: Efficiency inPublic Enterprise, Pub. Choice, 73, pp. 205–39.

Wallsten, Scott J. 2000a. “An Econometric Analy-sis of Telecommunications Competition, Priva-tization, and Regulation in Africa and LatinAmerica,” J. Indust. Econ. forthcoming.

———. 2000b. “Telecommunications Privatizationin Developing Countries: The Real Effects ofExclusivity Periods,” working paper, Stanford U.

Wasserfallen, Walter and Stefan Müller. 1998.“Deregulation and Privatization: Evidence fromthe Telecommunications Industry in Europeand Implications for Switzerland,” working pa-per, Studienzentrum Gerzensee, Switzerland.

Weiss, Andrew and Georgiy Nikitin. 1998. “Effectsof Ownership by Investment Funds on the Per-formance of Czech Firms,” working paper, Bos-ton U.

Welch, Ivo. 1989. “Seasoned Offerings, ImitationCosts, and the Underpricing of Initial PublicOfferings,” J. Finance 44, pp. 421–49.

388 Journal of Economic Literature, Vol. XXXIX (June 2001)

Wellenius, Björn. 2000. “Extending Telecommuni-cations beyond the Market: Towards UniversalService in Competitive Environments,” in Pri-vate Sector 21, World Bank, pp. 5–18.

Wolfram, Catherine D. 1998. “Increases in Execu-tive Pay Following Privatization,” J. Econ.Manag. Strat. 7:3, pp. 327–61.

World Bank. 1995. Bureaucrats in Business.Washington, DC: Oxford U. Press for the WorldBank.

Wurgler, Jeffrey. 2000. “Financial Markets and theAllocation of Capital,” J. Finan. Econ. 58, pp.187–214.

Xu, Xiaonian and Yan Wang. 1997. “OwnershipStructure, Corporate Governance, and Firms’Performance: The Case of Chinese Stock Com-panies,” working paper, World Bank.

Yarrow, George. 1986. “Privatization in Theoryand Practice,” Econ. Pol. 2, pp. 324–64.

Yergin, Daniel and Joseph Stanislaw. 1998. TheCommanding Heights: The Battle between Gov-ernment and the Marketplace That Is Remakingthe Modern World. NY: Simon & Schuster.

Yotopoulos, Pan A. 1989. “The (Rip)tide of Privati-zation: Lessons from Chile,” World Develop.17, pp. 683–702.

Megginson and Netter: From State to Market 389