You are on page 1of 20

Singapore Inc. versus the Private Sector: Are Government-Linked Companies Different? Author(s): Carlos D.

Ramrez and Ling Hui Tan Source: IMF Staff Papers, Vol. 51, No. 3 (2004), pp. 510-528 Published by: Palgrave Macmillan Journals on behalf of the International Monetary Fund Stable URL: http://www.jstor.org/stable/30035960 . Accessed: 26/06/2013 11:06
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Palgrave Macmillan Journals and International Monetary Fund are collaborating with JSTOR to digitize, preserve and extend access to IMF Staff Papers.

http://www.jstor.org

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

IMF StaffPapers Vol.51, No.3 c 2004 International Fund Monetary

the Private Singapore Inc.Versus Sector: AreGovernment-Linked Companies Different?


TAN* D. RAMIREZ and LING HUI CARLOS
Government-linked companies (GLCs)have a significantpresence in Singapore's corporate sector. Unlike parastatals in many other countries, these companies are run on a competitive,commercialbasis, ostensibly withoutgovernmentprivileges. Based on data from publicly listed GLCs and non-GLCs, we indeedfind no evidence that GLCs have easier access to credit. However, we do find that being a GLC is rewardedin financial marketswith a positive premium,over and above what can be explained by the usual determinantsof Tobin'sq. [JELL32,

L33, G32]

industrialization s partof its postindependence govplan,the Singapore role by establishingstate ernmentassumeda proactiveentrepreneurial or GLCs)in key sectorssuch (calledgovernment-linked companies, enterprises andservices.In as manufacturing, finance,trading, shipbuilding, transportation, was differentfrom Hong Kong SAR, whose economic this respect,Singapore andotherEastAsian economieslike growthwas drivenby privateenterprises, Japan,TaiwanProvinceof China, and the Republicof Korea,where active of enterprises. industrial ownership government policydidnotinvolvewidespread has been quitesuccessful. of "statecapitalism" this strategy By most accounts, andthe majorcompanational GLCshaveevolvedinto an important institution, andeven-in the names nies havebecomewell-recognized regionally corporate case of Singapore Airlines-globally.
*CarlosD. Ramirez is AssociateProfessor of Economicsat GeorgeMasonUniversity. Ling Hui Tan is a SeniorEconomist in the IMFInstitute. The views expressed in this paperarethose of the authors and do not necessarily thoseof theIMFor IMFpolicy.Theauthors thank Andrew Robert Feltenstein, represent at the IMFInstitute for helpfulcomments. Flood,CelineSia, andseminar participants

510

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

SINGAPORE INC, VERSUS THE PRIVATE SECTOR Unlike parastatals in many countries,Singapore'sGLCs bear a close resemblance to privateenterprises. The governmentsubscribesto whathas been termedthe in view" the "managerial ongoing debateon publicversusprivateownership,which that rather thanownershipper se is the key to efficiency.' GLCs argues competition are runon a commercialbasis, with a focus on bottom-lineperformance. They have not been used for social or employment-generation purposes.They compete with privatefirmsandmultinational companiesand,in some cases, with each other.Many of them have been partiallyprivatizedand are listed on the local stock exchange. Yet, while GLCs have undoubtedlybeen a majorelement in Singapore'seconomic development, there recently has been an active debate concerning their future role in the economy.2 Criticism of GLCs falls into two broad categories. The first contends that GLCs tend to do better than private sector firms because their institutionalrelationshipwith the governmentgives them special advantages in terms of access to funds, tenders, and opportunities;consequently, they have closed large areas of the economy to the private sector and stifled entrepreneurship. The second contends that GLCs tend to do worse than private sector firms because their managersare mainly civil servantswho lack business acumen, and their investmentsmay be politically, ratherthan commercially,motivated. The purposeof this paperis to investigatethe differencesbetween GLCs and privatesector firms empirically.To our knowledge, this has not been attemptedso far;the discussion has been mostly anecdotal.Using data on publicly listed GLCs and a control sample of non-GLCs,we considertwo questions:(1) Do GLCs benefit from special financial advantages?and (2) Do the financial markets value GLCs and non-GLCsdifferently,and if so, why? We examine the first claim by comparingthe investmentbehaviorof GLCs and non-GLCs-if GLCs do indeed receive preferential than financing,they would tend to be less liquidity-constrained non-GLCs.Next, we runTobin'sq regressionsto find out if government ownership/ affiliationmakes a difference to the marketvaluationof a company.

I. Background
To jump-startindustrialization in the late 1960s, the Singapore governmentcreated GLCs and statutory boardsto spearheaddevelopmentin varioussectorsof the economy.3The statedrationalefor this strategywas to compensatefor the lack of
SThealternative view, termedthe "political view" or "ownership thatgovernmentview,"maintains ownedenterprises areintrinsically inefficient becausegovernments in addition to, andin pursue objectives conflictwith,profitmaximization, andthis politicalinterference can distortthe objectivesandconstraints facedby managers of suchenterprises. See Shleifer(1998) for an exampleof this view, andBardhan and Roemer(1992) for an exampleof the managerial view. 2Thisdebatewas sparked committee on the role by the releaseof a report by a government-appointed of the government in business(Singapore, EconomicReviewCommittee, finalrec2002);thecommittee's ommendations are summarized in Singapore, Ministryof TradeandIndustry (2003). An earlierdebatein the late 1980s,whenseveralpublicenterprises were reflectingmanyof the samearguments-occurred fully or partially privatized. boardsarelegislatedunderindividual acts of Parliament thatdefinetheirfunctions,scope, 3Statutory and powers;they are formedundervariousministriesand are accountable to them throughParliament. GLCsareincorporated underthe Companies Act anddo not come underthe directpurviewof Parliament. Both GLCsandstatutory boardscan formtheirown subsidiaries andassociated companies.

511

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

Carlos D, Ramirezand LingHuiTan privatesector funds or expertise.PioneerGLCs includedthe Keppel, Sembawang, and Jurong Shipyards,which spurredthe development of Singapore as a major shipbuildingand ship repaircenter; the Development Bank of Singapore, which was set up to provide developmentfinancing; and Neptune Orient Lines, which was formedto leverageon the island'sstrategic location.Some GLCswere set up for Industriesand Allied Ordnancein the defense strategicreasons,notablyChartered industry. Many of these early companieswerejoint ventureswith foreigninvestors. For example,the SingaporeRefiningCompany,which providedthe catalystfor the growth of the oil refining industry,was a joint venture with Caltex and British Petroleum, while the PetrochemicalCorporationof Singapore, which launched was a joint venturewith Shell and Singapore's entryinto the petrochemicals industry, a Japaneseconsortium. In 1974, the government(throughthe Ministryof Finance) establisheda limited holding company,TemasekHoldings, to manage its investmentsin GLCs.At that time, 36 companies were transferred to Temasek'scontrol.4Since then, rapid economic growth has affordedGLCs the scope and opportunitiesto expand and of a diversify their operations.The 1980s and 1990s also saw the corporatization numberof statutoryboardsinto GLCs.5Today,the total numberof GLCs is estimated to be in the hundreds.TemasekHoldings directlyholds 22 first-tierGLCs, all of which have subsidiariesor associate companies, which in turn often have third-tiersubsidiaries,and so on. The companies are involved in a wide range of sectors, including finance, telecommunications,transportand logistics, property, and engineering,and utilities. infrastructure Temasek Holdings and its subsidiaries are registered companies under the as privatebusinesses. Many CompaniesAct, subject to all the same requirements of its companiesare listed on the SingaporeExchange.Accordingto Temasek,the major listed companies account for more than 20 percent of the total market capitalization.6 In additionto GLCs held directlyor indirectlyby TemasekHoldings, thereare also a numberof enterprisesthat are fully or majorityowned by statutoryboards. Such enterprisesmay also be classified as GLCs, to the extent thattheir sharesare owned ultimately by the government.An example is the Comfort Group, a publicly listed land transportationservices conglomerate owned primarily by the SingaporeLaborFoundation(a statutoryboard). According to the government,GLCs operate fully as for-profitcommercial entities, on the same basis as privatesector companies:They are expected to pro4Twootherholdingcompanieswere set up aroundthe same time: MND Holdings(ownedby the and Sheng-LiHoldings(ownedby the Ministryof Defense). MND Ministryof NationalDevelopment) to of Financeandthe bulkof its GLCstransferred takenoverby the Ministry Holdingswas subsequently for defense-related GLCs. is responsible Temasek. Technologies) Sheng-LiHoldings(now Singapore 5For example,the Telecommunications Authorityof Singaporewas convertedto SingaporeTelein 1992, the PublicUtilitiesBoard'selectricitysectorfunctionswere spunoff to create communications in 1997. wasconverted intoPSA Corporation of Singapore Powerin 1995,andthePortAuthority Singapore areDBS Bank,KeppelCorporation, 6Thelistedfirst-tier Lines,SembCorp NeptuneOrient companies See http:// and SingaporeTelecommunications. Industries,SingaporeAirlines, SMRT Corporation, www. temasekholdings.com.sg/.

512

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

SINGAPORE VERSUS THE PRIVATE SECTOR INC, vide commercialreturns,commensuratewith risks taken; they are subject to the sameregulations andmarketforces as privateentrepreneurs; andthey do not receive subsidies or treatment from the government.In an early surveyof any preferential in Singapore,Lee (1976, p. 57) concludesthat"[g]overnment conpublicenterprises trol is in fact very loose" and thatthe government"normally does not interferewith the managementof the companies directly."Lee (1976, p. 58) furthernotes that GLCs appearto receive few, if any, special privilegesby virtueof theirgovernment ownership:
Taxholidaysandtax concessionsare appliedgenerallyto all companies as long as theyfulfill the conditionsof pioneerstatusandexportorientation. Most of the publicenterprises obtaincreditfromthe Development Bankof Singapore, butthe interestrateis usuallynot lowerthanthatof other banks.... A few governmentcompaniesmay secure orders of A few saidthey might goods andservicesfromgovernment departments. becausethey squeezea lowerprice from suppliersof inputof materials were government-owned. ... To the question,"whether they can obtain the government's futurepolicy measures," priorinformation regarding all of themrepliednegatively. practically

The main advantageof governmentownershipappearsto be the positive signal it sends to the markets. The following statement by a GLC manager, cited in

Low (1991, p. 65), sums it up: "Being linked to Governmentis of course useful. It gives the companycredibilityand nobody will thinkyou are a fly-by-nightoperation. But the company has to justify itself and earn its keep by marketingright productsat the right time as no favoursare given or expected." Indeed, many GLCs have consistently posted a strongfinancialperformance. But the rapidgrowth of GLCs-both in size and in number-has led to concerns that they are encroachinginto too many industries,effectively crowding out the private sector and hinderingthe development of a critical mass of thrivinglocal enterprises.Among small- and medium-sized private enterprises in particular, GLCs are still perceivedto have unfairadvantagesin termsof access to funds, tenders, and opportunities.Other critics argue that GLCs are less efficient than private sector firms, due to their institutionalrelationshipwith the government,the marketstructure in which they operate,or the managementsystems appliedwithin them. For example, GLC managersare usually appointedfrom the ranksof senior civil servantsand militaryofficers; and, while they are generallyof a high quality and promoted on the basis of their performance-the Singaporeancivil service being "an extreme example of a meritocracy"(Krause, 1987, p. 119)-they have also been criticizedfor being too risk-averseand lacking sufficiententrepreneurial drive.7Therehave also been chargesthatcertainGLCinvestmentshave been politically ratherthan commercially motivated.And being linked to the government may sometimes be a hindranceratherthan an advantage;in recent years, some
withnewsenior years,someof thelargeGLCssawmanagement executives 7Inrecent shake-ups, in fromtheprivate sector andsomefromabroad. A number of theseforeign executives subsebrought leftbefore their contracts. Thegovernment hasputthisdownto coincidence, quently maincompleting that it doesnotplaya direct rolein recruitment decisions made boards. taining by theGLCs'

513

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

Carlos D. Ramirezand LingHuiTan countrieshave been reluctantto allow SingaporeanGLCs to invest in sectors considered nationallystrategic.8

II.Data and Descriptive Statistics


There is very little publicly availableinformationon GLCs in Singapore.Temasek Holdings lists its majorcompanieson its website, but not all of its (numerous)subsidiariesand associatedcompanies.In fact, the definitionof a GLC has itself been the sourceof some controversy. of Statisticsdefines GLCs Singapore'sDepartment as companies in which the government'seffective ownershipof voting shares is 20 percentor more.9However,as arguedin United States,StateDepartment (2001), this definitionexcludes many second- or third-tier GLC subsidiaries.For example, if Temasekowns 50 percentof a first-tierGLC, and that GLC owns 30 percentof a subsidiary,the effective governmentownershipof the subsidiaryis calculatedto be only 15 percent;as a result,the subsidiaryis not consideredto be a GLC. It can be arguedthat companies in which the government'sformal shareholdingis less even if they are not than 20 percentshould still be consideredgovernment-linked, For if the other shareholders individually effectively government-owned. example, own a miniscule fractionof the company,a mere 5 percentgovernmentownership can entailde facto controlof the company.Thus,the government may be the largest shareholder (and thereforehave a controllingstake) with an ownershipfractionof less than 20 percentif the other shareholders are atomistic. For the purposesof this paper,we classify a firmas a GLCif one of its substanboard.(Underthe Companies tial shareholdersis TemasekHoldingsor a statutory in a companyif he Act, an individualis consideredto have substantial shareholding Firm or she has an interestin 5 percentor moreof the votingsharesof thatcompany.) were obtainedfrom the Corporate data, including informationon shareholders, Handbook, supplemented,where necessary, by informationfrom the Singapore Exchange's website. A list of GLCs and the controlgroupof privateenterprisesis presented in Table 1. We focus on only three sectors-manufacturing; transport, storage, and communications (TSC); and multi-industry-as these are areas in We were able to obtain adequatedata which GLCs have a significantpresence.10 for the period 1994-98 for a sampleof 17 GLCs and 92 privateenterprisesin these three sectors. for each statisticsfor the sample.The figurespresented Table2 presentssummary deviations,calculated group(GLCand non-GLC)aremeans,medians,and standard for all firms and all years.Note thatthe averageGLC is almost 10 times as largeas the averagenon-GLCin termsof capitalstock(fixed assets).Aside fromthis, the two During the sample groups of firms have a roughly similar set of characteristics. cash flow and lower GLCs average gross sales (proportional reported slightly period,
in takeover in HongKongSAR Telecommunications was defeated attempts 8In 1999-2000,Singapore andMalaysialargelyfor this reason. was not to this definition 9See Singapore, of Statistics(2001).Thelist of GLCsaccording Department published. of in the financialand property 10GLCs are also stronglyrepresented sector,but the characteristics framework. firmsin these sectorsdo not lend themselveswell to ourempirical

514

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

INC,VERSUS SINGAPORE THE PRIVATE SECTOR

Table 1. Sample Companies


GLCs ComfortGroup CWTDistribution DelGroCorp. Intraco JurongShipyard KeppelCorp. KeppelHitachiZosen KeppelMarineIndustries NatSteel NeptuneOrientLines SembCorp Logistics Airlines Singapore Petroleum Co. Singapore PressHoldings Singapore Telecoms Singapore SNP Corp. TimesPublishing Sector TSC TSC TSC Multi-industry Manufacturing Multi-industry Manufacturing Manufacturing Multi-industry TSC TSC TSC Manufacturing Manufacturing TSC Manufacturing Manufacturing Private Enterprises Ace Dynamics AcerComputer International Acma & Devt AllianceTechnology AmtekEngineering Industrial Armstrong Corp. Asia PacificBreweries AvimoGroup AztechSystems BergerInternational BritishAmerican TobaccoCo.(S) Industrial Broadway Group BurwillHoldings CAMInternational Holdings CarnaudMetalbox Asia CerebosPacific ChuanHupHoldings ClipsalIndustries (Holdings) CompactMetalIndustries Cosco Investments (S) Creative Technology Datapulse Technology Eastern Publishing Elec & EltekInternational Co. EltechElectronics Falmac FirstEngineering Fraser& Neave LinksExpressHoldings Freight FuYuManufacturing GB Holdings General Magnetics GikenSakata(S) GP Batteries International GPEIndustries HawParCorp. HawParHealthcare HBMPrint HesheHoldings Ho WahGentingInternational HotelProperties HwaHongCorp. HwaTatLee Holdings IMCHoldings Inter-Roller Engineering IPCCorp. JayaHoldings Cement Jurong Sector Manufacturing Manufacturing Multi-industry Multi-industry Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing TSC Manufacturing Manufacturing TSC Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing TSC Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Multi-industry Manufacturing Manufacturing Manufacturing Manufacturing Multi-industry Multi-industry Manufacturing TSC Manufacturing Manufacturing TSC Manufacturing
(continued)

515

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

Carlos D. Ramirez and Ling HuiTan

Table 1. (concluded)
GLCs Sector Private Enterprises LiangHuatAluminium LionAsiapac Metalock(S) MHEHoldings NationalKap NetworkFoodsInternational Nippecraft OmniMold OspreyMaritime PacificCanInvestment Holdings PacificCarriers PanPacificPublicCo. PCI Pentex-Schweizer Circuits PokkaCorp.(S) Powermatic DataSystems Prima QAF Rothmans Industries Holdings RotolSingapore SanTeh SeksunPrecisionEngineering Sime Singapore SingamasContainer Holdings Singatronics SM SummitHoldings Sunright SuperCoffeemixManufacturing MetalPrinting Superior Teckwah Industrial Corp. The Straits Co. Trading TIBSHoldings TongMengIndustries TotalAccess Communication PublicCo. Tri-MTechnologies (S) TuanSing Holdings UnitedEngineers UnitedIndustrial Corp. UnitedPulp& PaperCo. Venture Manufacturing (S) WassallAsia Pacific WBLCorp. Wepco Yeo HiapSeng Note:TSCis transport, storage,andcommunications. Sector Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing TSC Manufacturing TSC Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Multi-industry Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Multi-industry Manufacturing Multi-industry Manufacturing Manufacturing Manufacturing Manufacturing Manufacturing Multi-industry TSC Manufacturing TSC Manufacturing Multi-industry Multi-industry Multi-industry Manufacturing Manufacturing Manufacturing Multi-industry Manufacturing Manufacturing

516

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

THE SINGAPORE INC.VERSUS PRIVATE SECTOR

Table 2. Summary Statistics


Statistic Numberof firms Investment* Mean
Median Standard deviation

GLCs 17 0.14
0.10 0.22

Non-GLCs 92 0.14
0.04 0.39

Cashflow* Mean
Median Standard deviation Gross sales* Mean Median Standard deviation Liquid assets* Mean Median Standard deviation q Mean Median Standard deviation ROA Mean Median Standard deviation

0.31
0.23 0.38 2.58 1.44 3.98 0.87 0.45 1.01 1.83 1.44 1.21 6.06 5.43 7.69

0.44
0.20 1.16 4.15 2.24 7.42 0.87 0.28 1.78 1.65 1.44 0.85 3.91 4.35 10.62

ratio Debt-equity Mean


Median Standard deviation

0.91
0.61 0.94 3,558 844 5,815

0.97
0.71 1.71 922 141 5,281

Totalassets(S$ million)
Mean Median Standard deviation

Fixed assets(S$ million) Mean


Median Standard deviation

1,407
301 2,621

149
46 367

to fixed assets. *Proportional

to capitalstock)compared withnon-GLCs; andliquidassets(also averageinvestment to capitalstock) were almostidenticalfor the two groups.The medians proportional thattheirdistributions of thesevariables arealwayssmallerthanthe means,indicating are skewedby some largevalues.This is truefor both GLCsand non-GLCs. Table2 also presentsthe value of q, a proxy for Tobin'sq (the marketvalue of the firm relative to its replacementcost). Here, q is approximated by the market value of common equity stock plus the book value of debt and preferredstock,

517

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

Carlos D. Ramirezand LingHuiTan divided by the book value of total assets. During the sample period, the average Tobin's q for GLCs exceeded the average q for non-GLCs by about 10 percent, althoughthe medianq values were comparablefor the two groups.We investigate differences in Tobin'sq furtherin Section IV. Finally, Table 2 shows the profitabilityof the two sets of companies, as measured by the returnon total assets (ROA), which is the ratio of pretax net profits plus interestpaymentsto total assets (both equity and nonequitycapital). By this indicator,GLCs appearto be more profitableon averagethannon-GLCs.However, the standard deviationsare so largethatthe differencebetweenthe two meansis not levels.ll This can be verified by looking at the statisticallysignificantat standard medians,which are much more comparablebetween the two groups.Furthermore, this accounting ratio has well-known weaknesses-for example, it reflects only past profitability,is not adjustedfor risk, and tends to be subjectto manipulation. We discuss this furtherin Section IV.

LessLiquidity-Constrained? III.AreGLCs
A recurring chargeagainstGLCs is that they enjoy financialprivilegesand do not an equal footing with privateenterprises.Krause (1987, p. 119), for on compete advantage... in marshallingfinancial example, assertsthat GLCs have a "natural saves morethanit invests,andthus alwayshas because"thegovernment resources," easy access to finance."Chargesof cheap fundingfor GLCs were also made more debatelast year.12 duringa parliamentary recentlyby two membersof Parliament There are different ways in which GLCs could potentially receive cheap funding-government loan guarantees,concessional interest rates, and generous repaymentperiods are some examples that come to mind. However,it is virtually impossible to obtain data on the terms of borrowingfor each firm and each project. Hence, a directtest of preferentialcreditaccess is not feasible, and we have to resortto an indirecttest. One useful way to test-indirectly-if GLCs have indeed received financial advantagesis to comparetheir investmentbehaviorwith that of their privatesector counterparts. problemsof Accordingto the imperfectcapitalmarketsliterature, asymmetricinformationbetween borrowersand lenders(such as adverseselection and moral hazard)make externalfinancingmore costly thaninternalfinancingfor many firms. If lendershave imperfectinformationaboutthe qualityor riskinessof the borrowers'investment projects, adverse selection will lead to the so-called "lemons"premium,a wedge between the cost of externalfinancingand internally generatedfunds. In the presence of incentive problems and costly monitoringof managers, external lenders will require a higher returnto compensate them for these monitoringcosts and the potential moral hazardassociated with managers' (i.e., borrowers')controlover the allocationof investmentfunds. To the extentthat a firm'smanagersthemselves supplythe funds for investmentprojects,the shadow cost of these funds need not carrysuch a premiumreflecting adverse selection or
for the difference in meansis 0.99 for the ROA. The t-statistic 11 of Information 12SeeSingapore, (2002). Ministry

518

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

SINGAPORE INC. VERSUS THE PRIVATE SECTOR Firmsthatfind it too expensiveto raisecapitalexternallywould thus moralhazard.13 tend to rely more on internalfunds to finance their investmentspending, so their investmentspendingwould tend to be stronglycorrelatedwith their liquidity-all else constant,more liquid firms would be able to affordto invest more. If GLCs have preferentialaccess to financing (whether it be through lower interestrates or governmentguarantees),this means they find it cheaper to raise externalfunds comparedwith their privatesector counterparts. If this is the case, as a determinant of GLCs' liquidityshouldbe irrelevant(or at least less important) investmentspending.The test, therefore,involves comparingthe differencesin the effects of various determinants of investmentspendingby GLCs and privatesector companies:If GLCs have better access to credit, then their investmentspending shouldbe less sensitive to a liquiditymeasurelike cash flow thanwould be the case for their private-sector counterparts. The idea for this test is not new. Previousempiricalstudieshave determinedundera varietyof settings--thatfirmsthatwouldappear to face the greatest problems most in responseto cash flow shortraisingcapitalexternallytend to cut investment falls. For example,Fazzari,Hubbard, and Petersen(1988) use data fromAmerican firms,identifyingfirmsthatretaina smallerfractionof theirearnings manufacturing as being less liquidity-constrained; Hoshi, Kashyap,and Scharfstein (1991) use data from Japanesemanufacturing firms, identifyingfirms linked to a keiretsuas being less liquidity-constrained; and Ramirez(1995) uses datafromAmericancompanies in the 1910s, identifyingfirms affiliatedwith J. P. Morganas being less liquidityconstrained.14 More recently and closer to our subject, Harrisonand McMillan are (2001) use data from C~te d'Ivoireto test if foreign firms and state enterprises less credit-constrained thandomesticprivatefirms. The regression specification for these tests is essentially a reduced-form investmentequation,with gross investmentregressedon the standard explanatory variables, q and gross sales, as well as on liquidity measures such as cash flow. Our specificationis based on thatused in Hoshi, Kashyap,and Scharfstein(1991), and takes the following form:
I,
=

l + Po + p1q<-1 + yi(GLC +73 GLC x


x

P211

311

CASH,
K,_1

(LIQ1
41 KK_1

Kt_1

K_1 SALESl,

SALES, + y2 GLC x SL

q,~-)
CASH,

K K1

Y4 GLC x

LIQ,

K_(1) K

Q_ + (1,.

13SeeHubbard and a graphicalanalysisillustrating (1998) for a review of this literature the link betweeninternal fundsandcapitalinvestment in modelsof informational asymmetries. andZingales(1997) disputethe claim thatcapitalmarket arecorrelated with '4Kaplan imperfections investment-cash flow sensitivities.However, theirresultsare subjectto criticismon bothconceptual and Forexample,theirconclusionsarebasedon a very smallsampleof firms,whichthey empirical grounds. thensubdivide into five categories to clearlyendogenous variables suchas the degreeof liquidaccording andPetersen(2000) for a criticalreply.See also HennessyandLevy (2002) for Hubbard, ity. See Fazzari, a morerecentoverviewof this issue.

519

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

Carlos D. Ramirezand LingHuiTan The dependentvariableis investmentduringperiod t (I,) as a proportionof capital stock at the beginningof period t (K,/1). The independentvariablesare average Tobin'sq at the beginningof period t (qt-1); lagged gross sales (SALES,_1)relative to capital stock; cash flow during period t (CASHt)relative to capital stock; the stock of liquid assets at the beginningof period t (LIQ,_1) relativeto capital stock; and the same four variablesinteractedwith a dummy(GLC),which takes the value of 1 for GLCs and 0 otherwise.15 The motivationfor this regressionis set out in Hoshi, Kashyap,and Scharfstein (1991) and also discussed in detail in Hubbard(1998); we briefly review the general idea behind it below. Neoclassical investment theory predicts that a firm's investment spending shoulddependpositively on its investmentopportunities, thatis, the expectedpresent value of futureprofits from additionalcapital investment.This expectationis capturedby the value of marginalq, the shadowvalue to the firm of an additional we use unit of physical capital.As a proxy for marginalq, which is unobservable, averageTobin'sq at the beginningof each period,qt-1.16The empiricalinvestment literaturefinds that investmentspending is also positively correlatedwith lagged effect. We use lagged gross sales, SALESt1,to proxy for outputvia the accelerator as an explanatoryvariableto capturethis effect. In the absence of lagged output friction, these two variablesshould be sufficientfor explaininginvestmentspending behavior. The imperfect capital markets theory extends these conventional a role for financingconstraintsin determining models of investmentto incorporate investment.As explainedabove, models of asymmetricinformationand incentive costs and the internalresources problemsin capitalmarketsimply thatinformation of a firm influence the shadowcost of externalfunds for fixed investment,holding constantunderlyinginvestmentopportunities. Simply put, the theorypredictsthat, all else being equal, investmentshould be significantlycorrelatedwith the change in net worth (internalfunds) for firms that are likely to face information-related The standard proxyfor liquidityis cash flow (CASHt), capitalmarketimperfections. income after tax that is, plus (accounting)depreciationless dividendpayments.17 Following Hoshi, Kashyap,and Scharfstein(1991), we also include a stock meaassets such as marsure of liquidity,thatis, the firm's stock of cash and short-term ketable securitiesat the beginningof the period (LIQt-l)-these are assets thatcan be readily convertedto cash to finance investmentspending. One potential concern with the regression specification is that cash flow could be correlated with other determinantsof investment, such as expected
for GLCsand 15Analternative wouldbe to estimatethe investment regressions separately approach in the differin equation(1) becausewe are interested non-GLCs. However,we preferthe specification of investment ence in investment behaviorbetweenGLCs and non-GLCs,ratherthanthe determinants spendingfor each groupof firmsper se. (1998), this is allowableundercertainassumptions, including perfectcompe16Asnotedin Hubbard of production of fixed capital,linearhomogeneity titionin the factorandproduct markets, homogeneity andinvestment decisions. of financing costs, andindependence technologiesandadjustment (1998), cash flow is an imperfect proxyfor the changein net worth.For 17Asdiscussedin Hubbard decisions(timingandfinancial) of a firm'scash flow may reflectaccounting example,the determination in manycases, it is the best available with the changein net worth.However, thatmuddyits correlation proxy.

520

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

SINGAPORE INC. VERSUS THE PRIVATE SECTOR future profitability or sales. If this is the case, a link between cash flow and investment for a given firm over time could be reflecting the link between expected profitability and investment, and a cross-sectional link between cash flow and investmentat a point in time could be reflecting the fact that firms with high cash flow have successful investments or low costs and face incentives to expandproduction.However,the inclusion of Tobin'sq and gross sales as regressors should control for these factors, as the component of shifts in net worth accounted for by changes in currentand expected future profitability should be capturedin gross sales and q. Of course, the average (Tobin's) q that we use may be a poor proxy for marginalq, the theoreticalconstructthatis a measureof investmentopportunities, so anotherpotential problem with the regression is the (mis)measurementof q. However,the point of our regressionis to comparethe cash flow and liquid assets coefficients for the two sets of firms. This difference should be unbiased as long as the mismeasurement is the same for both sets of firms. A third source of concern could be selection bias-selection bias would be a problemif there is endogeneity between GLCs and liquidityconstraints,that is, if the governmentsomehow chose only less-liquidity-constrained firms to be GLCs. But as noted earlier,many GLCs were created in the late 1960s and early 1970s as part of the government'sstrategyto spearheaddevelopmentin various sectors of the economy; the GLCs in our sample were mostly incorporatedduring that early period.It would be difficultto arguethatthe governmentcould perfectlypredict the performanceof these GLCs at the time of their creation. Our sample of GLCs does not containany instanceof the governmentcherry-picking privatesector firms and convertingthem into GLCs. As the GLCs in our sample are historically determined,then, for the purposesof statisticalinference duringthe sample period, we treatbeing a GLC as an exogenous event. Equation (1) is estimated using ordinary least squares (OLS) with fixed effects. (Firmand year dummieswere includedin each regression.)The regression results, shown in Table 3, column 1, indicate that the liquidity variables-cash flow and the stock of liquid assets-are indeed importantexplanatoryvariables. The other explanatory variables also have the right signs and are statistically significant. As our focus is on the difference between GLCs and non-GLCs, we are interested in the coefficients on the interaction terms involving the liquidity variables. If GLCs are less liquidity-constrainedthan non-GLCs in their investment decisions, then the coefficient on GLC x (CASH, / Kt-1) and GLC x

(LIQt-1/ Kt-1) should be negative and significant. Table 3, column 1 shows that the GLC cash flow coefficient is smallerthanthe non-GLCcash flow coefficient~as evidenced by the negative coefficient on the interactionterm, GLC x (CASHt/ K/_1)-but the difference between the two cash flow coefficients is not significant. The GLC liquid assets coefficient is larger than the non-GLC cash flow coefficient-as evidenced by the positive coefficient on the interaction term, GLC x (LIQt- 1 / K/_l)-but again, the difference between the two coefficients is not significant. Hence, our tests show no statistical difference between the GLC and non-GLC liquidity coefficients. This indicates that the GLCs in our 521

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

Carlos D. Ramifrezand Ling HuiTan

Table 3. Regression Results: Liquidity


Dependent Variable: It/ K-1 / K,-1 CASH, LIQ,- / K,-1 / K,-1 SALESt-I qt-1 GLCx (CASH, / K,-1) GLCx (LIQ,t1 / K,-1) GLCx (SALES,_1 / K,_t) GLCx qt-1 Number of observations Degreesof freedom R2(within) R2(between) R2(overall) (1) Full Sample (2) Multi-industry (excluding small non-GLCs) 0.195 (0.274) 0.222* (0.087) -0.020 (0.050) -0.064 (0.277) -0.029 (1.047) 0.022 (0.186) -0.012 (0.074) 0.129 (0.758) 51 30 0.3513 0.0002 0.1002 (3) TSC (excluding small non-GLCs) -0.279 (0.178) 0.254 (0.295) 0.257 (0.168) 1.228" (0.375) -0.185 (1.426) -0.104 (0.568) 0.237 (0.467) -1.083* (0.405) 61 37 0.4263 0.0020 0.0237 (4) Manufacturing (excluding small non-GLCs) 0.115 (0.100) 0.093 (0.059) 0.097* (0.019) 0.111" (0.046) -0.021 (0.431) 0.001 (0.244) -0.011 (0.298) -0.121 (0.309) 135 92 0.3948 0.0287 0.0801

0.142" (0.051) 0.082* (0.022) 0.018" (0.007) 0.124" (0.037) -0.143 (0.319) 0.125 (0.116) -0.034 (0.032) -0.082 (0.096) 421 304 0.1833 0.0196 0.0312

Notes: Firmand year dummiesare included.Standard errorsare presented in parentheses. An asterisk(*) indicatessignificance at the 5 percentlevel (two-tailed test).

sample are no more or less liquidity-constrainedin their investment decisions than their private sector counterparts.Our results are therefore consistent with the government's claim that GLCs do not enjoy cheap funding because of their link to the government. The fact that we fail to find significant coefficients on the two interaction terms of interest may be due to the disparity in sample sizes (17 GLCs versus 92 non-GLCs)or, more specifically,to the relativelysmall numberof GLCs in our these 109 firms representthe limit of the data providedin sample. Unfortunately, the (comprehensive)CorporateHandbook.Breakingup or otherwise manipulating the samples does not change our results. To illustrate,columns 2, 3, and 4 in Table 3 present the results of the same regression run separatelyby sector and excluding the smallest non-GLCswith averagefjxed assets below S$50,000. The 522

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

SINGAPORE VERSUS THE PRIVATE SECTOR INC,


estimated coefficients on GLC x (CASH, / Kt-1) and GLC x (LIQ-1 / Kt-1) are not

significantlydifferentfrom zero in each case. Our findings are, of course, based only on the sample period 1994-98.

one as it However,while relativelyshort,this periodis actuallyan important


includes the years just after the onset of the Asian crisis. One can conjecturethat if GLCs are eligible for special treatment,they will be more likely to need such favors when times are bad (e.g., during the Asian crisis) than when times are good. The fact that we find no difference in liquidity constraintsbetween GLCs and non-GLCsduringthis period is thus a strongrejection of the hypothesis that GLCs have easier access to credit.18 A second, more important,caveat is that our relate to GLCs that have been partiallyprivatizedand are findings obviously only publicly listed; unlisted GLCs or wholly owned governmentcompanies may well behave differently.

IV.Does the Market and Non-GLCs Value GLCs Differently?


If GLCsdo notseemto be benefiting fromspecialfinancial doesbeing privileges, Inc. makeany difference at all? Havingfoundno evidenceof partof Singapore creditaccess for GLCs,we turnour attention in this section to a preferential
different question, namely, how does the marketperceive GLCs compared with non-GLCs? As mentionedearlier,some GLCs claim thattheironly advantageis thatbeing linked to the governmentsends a positive signal to the markets(much like bearing a "Good Housekeeping stamp of approval"). In this section, we investigate whetherthe market'svaluationof a firm is correlatedwith its governmentlink. We have no prior beliefs about whether the governmentlink would translateinto a higher q, lower q, or no effect on q. If GLCs differ from non-GLCsbecause the government link enhances performance, for example, then one would expect GLCs to be valued more than comparablenon-GLCs. On the other hand, if the capital marketsperceive GLC managers to be corruptand inefficient, then one would expect the governmentlink to translateinto a lower q. Following the approachof Lang and Stulz (1994), we focus on Tobin's q ratherthan on stock marketor accounting measures of performancein order to avoid some of the problems associated with those comparisons. The numerator of q-the firm's marketvalue-reflects the firm's expected future profits, while the accountingrate of returnmeasures only past profits. Furthermore, the firm's marketvalue also incorporatesthe varianceof expected profits, so q includes an automaticadjustmentfor risk; by contrast,comparisonsof stock returnshave to account for differencesin risk. Of course, the underlyingassumptionsbehind the use of Tobin's q are that financial marketsare efficient and that a firm's market value is an unbiasedestimate of the presentvalue of its cash flows.
the time frameof the analysisso as to compare 18Apossible extensionwould involve lengthening resultsduringthe pre-Asian crisisperiodwith the post-Asiancrisis period.The drawback hereis thatthe as severalcompanies(includingsome gain in lengthof the dataset wouldbe offset by a loss in breadth, GLCs)do not havea very long historyof being listedon the stockexchange.

523

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

Carlos D. Ramirez and Ling HuiTan

Table 4. Regression Results: Tobin's q


In q, variable: Dependent GLC (Earnings / Price) Assets) In (Total (Debt/ Equity) ROA Beta dummies Industry Yeardummies Number of observations R2 Adjusted No Yes 530 0.1659 Yes Yes 530 0.2789 Yes Yes 530 0.3815 Yes Yes 530 0.3804 Yes Yes 530 0.3811 Yes Yes 529 0.4529 (1) 0.085 (0.103) (2) 0.226* (0.090) (3) 0.224* (0.089) 0.873* (0.090) (4) 0.215" (0.110) 0.868* (0.089) 0.006 (0.030) (5) (6) (7) 0.238* (0.101) 0.535* (0.218) -0.021 (0.027) -0.000 (0.015) 0.014* 0.045 0.217*" 0.199" (0.110) (0.095) 0.832* (0.089) 0.006 (0.030) -0.013 (0.017) 0.524* (0.218) -0.015 (0.025) -0.002 (0.014) 0.014*

(0.004) (0.005) (0.060)


Yes Yes 464 0.4550

errors arepresented in parentheses. An asterisk Notes:Robuststandard (*) indicatessignificance at the 5 percentlevel;a doubleasterisk(**) indicatessignificance at the 10 percentlevel (two-tailed of columnnumbers. test). See text for explanation dummies:(1) machinery and equipment, (2) electronicproducts,(3) metal products, Industry and andplastic,(5) food andbeverage,(6) chemicalproducts, (4) rubber (7) electrical,(8) printing (12) trans(9) transport (10) petroleum (11) othermanufacturing, publishing, equipment, products, port,(13) storage,(14) post andtelecommunication.

Table4 reportssemilog OLS regressionsof q on the GLC dummyand several control variables.Theory does not dictate a specific functionalform for a q equation, althoughHirschand Seaks (1993) provideevidence thatthe semilog form is superior.The semilog specificationalso has the advantageof dampeningthe influence of extreme or mismeasuredvalues of q.'9 The first column of Table4 shows that,controllingonly for year effects, In(q) for GLCs is higher than that for non-GLCsby 0.085 on average.This is more or less in line with our earlierobservation, basedon a comparison of meansin Table2. However,the difference is not statisticallysignificant. Column 2 refines the analysis by including 14 industry dummies into the We find thatindustryeffects regression(the base categorybeing "multi-industry"). increase the magnitudeof the GLC premiumfrom 8.5 percentto about 22.5 percent, and the coefficient on the GLC dummy is now significant at the 5 percent level. This suggests thatGLCs tend to be in low q industriesbut tend to have high
the regressions model(regression on groupmeanswith'9Wealso estimated usinga between-effl'ects out the yeardummies),andfoundthe resultsqualitatively unchanged.

524

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

SINGAPORE INC, VERSUS THE PRIVATE SECTOR q's relative to their industry cohort.20How do we explain this difference in the marketvaluationof GLCs comparedto non-GLCs? One possible explanationis that GLCs are more likely to operatein protected marketsand have a certaindegree of monopoly power.If GLC stocks were selling for higher prices on the Singapore Exchange because these companies were exploiting monopoly power, we would expect GLCs to have a high earnings-price ratio-earnings in the present would be high, but marketvalue (price) would not rise in proportion because investorswould look forwardto the long-runerosion of monopoly power in the face of new entry-and thatinclusion of the earnings-price ratio would reduce the GLC coefficient. However,this does not appearto be the case: the thirdcolumn shows that the GLC coefficient is not affected by the inclusion in the regressionof the companies' earnings-priceratios. Could the GLC premiumbe reflecting a size effect and economies of scale? As noted earlier,GLCs tend to be big companies:the averageGLC in our sample is about 10 times largerthanthe averagenon-GLC,in termsof total assets. Larger firms may be better able to exploit scale and scope economies. If this is the case, largerfirms would have a cost advantageover smallerones and may thereforebe more profitable.Column 4 considers the effect of size (measuredby the log of total assets) on q.21If the GLC premiumis due to their larger size, the inclusion of this variableshouldeliminate (or at least substantially reduce) the impactof the GLC dummy.22 We find that the GLC dummy does decrease by about 4 percent (from 0.224 to 0.215), but remainsstatisticallysignificant. A thirdpossibility is that GLCs may have lower debt-equityratios than nonGLCs. If GLCs are less leveragedthantheirprivatesectorcounterparts, theirprobof and failure would also tend to be and this lower, ability insolvency may explain theirhigherq. Column5 adds the debt-equityratioto the list of regressorsin order to controlfor the risk of bankruptcy. The results indicatethis variablecarrieslittle explanatorypower. More important,the coefficient on the GLC dummy is not materiallyaffected by its inclusion-it remainslarge and statisticallysignificant. Or perhapsGLCs are more profitablethan non-GLCs.This could be the case if they are better run or if they receive special discounts on inputs, for example. Column 6 shows the effect of adding the ROA to the regression. The estimated GLC coefficient declines in size by about 8 percent (from 0.217 to 0.199), but remains statisticallysignificant.This suggests that part of SingaporeInc.'s added value comes from making the GLCs more profitable.However,the fact that more

20That GLCsseemto be in low q industries actuallymaynotbe too surprising, giventhe factthatthey were established mostlyin areaswherethe privatesectorwas-at least initially-unwilling to go. In our suchas transport, andtransport sample,GLCsaremostlyfoundin industries storage,petroleum products, andgenerallyareabsentfrommanufacturing industries equipment manufacturing, involvingelectricaland electronicproducts, andequipment, metalproducts, chemicalproducts, etc. machinery 21 of size in order to control forpossiblenonlinearities (forexamWeuse thelog of assetsas ourmeasure as size increases). ple, the effectof size on q maydiminish 220ne may also posit, a priori,a negativecorrelation betweensize andq. Young,typicallysmall,and firmstendto havean expectedhighfuturegrowth promising profile,so q maytendto be higherfor smaller firms.In anycase, theimportant after pointis thatthe GLCdummyremains relatively largeandsignificant for size. accounting

525

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

Carlos D. Ramirezand LingHuiTan than90 percentof the premiumsurvivesthe inclusion of this variablesuggests that investors are also valuing the capitalized worth of future higher earnings from being affiliatedwith the government. Finally,we considerthe possibility thatGLCs' stock returnsmay have a lower covariancewith the marketreturn.If this is the case, it would imply that market movements have little influence on GLCs' returns,thereby making them more attractiveas investmentsin a portfolio that minimizes risk exposure. Column 7 shows the effect of addingeach firm'sbeta to the regression.Beta is the covariance between the firm's returnand the marketreturn,divided by the marketvariance, If investors'appetitesfor GLC thatis, it is the "normalized" covarianceof returns.23 the inclusionof beta shouldreducethe size stocks are drivenby this consideration, of the GLC premium.The regressionsresultsindicate,however,thatthis is not the case; instead,the inclusion of beta increasesthe premiumto 24 percent. In summary,comparingthe Tobin's q of GLCs and non-GLCs, we find evidence thatthe capitalmarketsvalue GLCs more highly thannon-GLCs.This positive and significant relation between the governmentlink and q is robust to the inclusion of other variables-such as industryeffects, size and monopoly power, profitability,and bankruptcyrisk-that might affect firm value and thereby q. Takingthese variablesinto considerationstill leaves us with a GLC premiumof more than 20 percent. Thus, performancemeasures aside, the capital marketsseem to rewardsubstantiallythe very fact that a company is linked to the government.This positive market perception is hard to pin down. It could simply reflect a form of brand recognition (much like how consumersare willing to pay more for goods bearing a well-known label than for similar or even identical goods without such a label). Or investorsmay believe-rightly or wrongly-that GLCs are backedby the government,which will not let them fail in times of trouble.(Krause, 1987, notes that a few small GLCs were permittedto fail in the 1970s and a few otherswere closed down in the 1980s, but the total capitalizationof these failures was relatively small.) V. Conclusion Singapore'sGLCs are an unusualbreed of state enterprises.Primarilyestablished to catalyze the industrialization process, they have expandedinto all areas of the economy, including those served by private enterprises.The governmentclaims that GLCs are run on commercial ratherthan ideological grounds, with no state interferenceor favors:They are expected to be efficient andprofitable; they are not supposed to receive special privileges or concealed subsidies; they are free to

23Betais obtainedfrom the CorporateHandbook,where it is computedusing the naturallog of of the All-Singapore of the stock againstthe weekly returns EquitiesIndexas the market weekly returns is 1. The indexfor the 105-weekperiodfromAugust29, 1997,to August27, 1999.The betaof the market returns of stocks with beta coefficientsgreaterthan 1 would,on average,respondto a largerextentthan as a whole. Conversely, those with betacothe market portfolioto factorsthataffectthatcapitalmarket to a lesserextent. efficientsless than I wouldrespond

526

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

SINGAPORE INC.VERSUS THE PRIVATE SECTOR

recruitstaff in the open market,both at home and abroad,on competitive terms; and they should be allowed to fail if they lose money.Yet time and again, charges of favoritismfor GLCs at the expense of privateenterprisessurface. Do GLCs receive special financingprivileges or not? This paperfinds no basis for the argumentthat GLCs have easier access to credit. Our regression results indicate that GLCs are no more or less liquidity-constrained in their investment decisions thantheirprivatesector counterparts. This suggests that GLCs are competing on a level playing field as far as access to financingis concerned. However,we do find that being a GLC is rewardedin financial marketswith a premiumof more than20 percent.This is afteraccountingfor the fact thatGLCs tend to be large (so that they may be able to exercise monopoly power or exploit economies of scale and scope), profitable, and less likely to go bankrupt.This GLC premiumhas to reflect the market'sperceptionof the benefits-whether real or illusory-of being linked to the government. While our resultsare interesting,they do not imply thatthe governmentshould keep creatingnew GLCsor expandingexisting ones. If the GLCpremiumis largely due to the market'sperceptionof the benefits of being linked to the government, further proliferationof GLCs-which will tend to stretch the resources of the government-will only dilute these perceivedbenefits and, thereby,the premium.

REFERENCES
and JohnE. Roemer,1992, "Market Socialism:A Case for Rejuvenation," Bardhan, Pranab,
Journal of Economic Perspectives (U.S.), Vol. 6 (Summer), pp. 101-16. Corporate Handbook Singapore 2000 (Singapore: CEIC Holdings Ltd.).

Fazzari, Steven, R. Glenn Hubbard,and B. Petersen, 1988, "FinancingConstraintsand Institution), pp. 141-95. "Investment-Cash Flow SensitivitiesAre Useful:A Commenton Kaplanand ,2000,
Zingales," Quarterly Journal of Economics, Vol. 115 (May), pp. 695-705. CorporateInvestment,"Brookings Papers on Economic Activity: 1 (Washington: Brookings

Ann E., and Margaret Harrison, Affect McMillan,2001, "Does Direct Foreign Investment Domestic Firms' Credit Constraints?" NBER WorkingPaper No. 8438 (Cambridge, Massachusetts: NationalBureauof EconomicResearch). A., andAmnonLevy,2002, "AUnifiedModelof DistortedInvestment: Hennessy,Christopher of California at Berkeley,Haas TheoryandEvidence" (unpublished; Berkeley:University School of Business). G. Seaks, 1993, "Functional Formin Regression Hirsch,Barry Modelsof Tobin's T., andTerry
q," Review of Economics and Statistics (U.S.), Vol. 75 (May), pp. 381-85.

Anil Kashyap, andDavidScharfstein, Hoshi,Takeo, and 1991,"Corporate Structure, Liquidity, Investment: EvidencefromJapanese Industrial Journalof Economics, Groups," Quarterly Vol. 106 (February), pp. 33-60. R. Glenn,1998,"Capital-Market andInvestment," JournalofEconomic Hubbard, Imperfections Literature (U.S.),Vol. 36 (March), pp. 193-225. Flow SensitivitiesProvide Kaplan,StevenN., andLuigiZingales,1997,"Do Investment-Cash Useful Measuresof FinancingConstraints?" QuarterlyJournalof Economics,Vol. 112 (February), pp. 169-215.

527

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions

Carlos D. Ramirez and Ling HuiTan Krause, Lawrence B., 1987, "TheGovernment as an Entrepreneur," in TheSingapore Economy Reconsidered, ed. by Koh Ai Tee, Krause,and Lee (Tsao)Yuan(Singapore: Instituteof Southeast Asian Studies). Lang,LarryH. P., and Rene M. Stulz, 1994, "Tobin's q, Corporate Diversification, and Firm Performance," Journalof PoliticalEconomyVol. 102 (December), pp. 1248-80. Lee, ShengYi, 1976, "PublicEnterprise and EconomicDevelopment in Singapore," Malayan EconomicReview, Vol. 21 (October), pp. 49-73. Low, Linda,1991, ThePoliticalEconomy ofPrivatisation in Singapore: Analysis, Interpretation, and Evaluation (Singapore andNew York: McGraw-Hill). Ramirez,Carlos D., 1995, "Did J. P. Morgan'sMen Add Liquidity?Corporate Investment, CashFlow,andFinancial Structure at the Turnof the Century," JournalofFinance,Vol. 50 (June),pp. ~61-78. Shleifer,Andrei, 1998, "StateversusPrivateOwnership," Journalof EconomicPerspectives, Vol. 12 (Autumn), pp. 133-50. Singapore, Department of Statistics, 2001, "Contribution of Government-Linked Companies to GrossDomesticProduct," OccasionalPaperon EconomicStatistics(March). Singapore, EconomicReviewCommittee, 2002, Recommendations on Goverrtmertt in Business (May 30). Avai lable via the Internet: http:~www. mti.gov. sg/public/PDF/CMT/ER C EISC MainRepO rt.pdfsid=I 13&cid=l 120. Singapore, Ministry of Information,2002, "Motion on Temasek Charter and EISC's Recommendations on Government in Business--SpeakingPoints for DPM's Round-up Speech,"SingaporeGovernment Press Release, August 28. Availablevia the Internet: http ~ivww. mti. gov. sg/public/fmzcmngrintpage. asp?SID=l 20&CID=l243&DID= &Page=. Singapore, Ministryof Tradeand Industry, 2003, "NewChallenges,FreshGoals--Towards a DynamicGlobalCity," Reportof the EconomicReviewCommittee, February 4. Available via the Internet: http:~www. mti. 80vo sg/public/PDF/CMT/ER C Comm Cover v2.pdf? sid=150&cid=1480. UnitedStates,StateDepartment, EmbassySingapore, 2001, "Government-Linked Corporations Face the Future,"Economic Report, March. Available via the Internet:http~www. usembassysingapo re. org.sg/epn001/GLC2000. html.

528

This content downloaded from 110.159.24.212 on Wed, 26 Jun 2013 11:06:13 AM All use subject to JSTOR Terms and Conditions