(31?21

Public Disclosure Authorized CFSDiscussion Paper Series

JapaneseNational Railways PrivatizationStudy II: Public Disclosure Authorized

InstitutionalizingMajor PolicyChange andEx.amimng Economic Implications

Authors

Public Disclosure Authorized Koichiro Fukui KiyoshiNakamura Tsutomu Ozaki Hiroshi Sakamaki Fumitoshi Mizutani Public Disclosure Authorized Cofinancing and Financial Advisory Services (CFS) Private Sector Development and PrivatizationGroup (CFSPS> TheWorld Bank December 1994 CFSDiscussion Paper Series, No. 107

Japanese National Railways Pnvatization Study II:

Institutionalizing Major Policy Change and Examining Economic Implications

Atwhors: Koichiro Fukui, Japan Development Bank Kiyoshi Nakamura, Professor, Waseda University Tsutomu Ozaki, Japan Development Bank Hiroshi Sakamaki, Japan Development Bank Fumitoshi Mizutani, Associate Professor, Kobe University

Foradditional copies of thispublicathn pleasecontact: Cofinancingand FinancialAdvkory Senvces. Room Q-5030. (202) 473-1228. Foreword

Ihe hardwork of rstructuring a statr-ownedenterprise and preparingit for privaxiztion goes on well afia the initi hecadyreorganization-which marks the beginning of the process-has ended. Even after finl planshave been developedand agreedto, a nced persistsfor flexibilityand fine uming in the tacil planning process,which proceedsin paralle with the implementationof rcstructuringof a large state enterprise.This volumeallows us to lookinside the processof enterpriseresrructuring. With unique clarity and insightit explainsthe responsesand resourcfailnessofJapanese authorities in continuingto guide one of te lagest smte-ownedenterprises-Japanese National Railways-through the thicketsand challngs ofprirarzation. It allowsus tomeasure theprogresswhhhas bcenmade sinceambitious prinati2aiion plans wereagreed to in 1986 and cxplainsthe ongoingneed for fine tuning of restructuringefFirts in responseto unanticipatedshifs in dte busines and competitiveenvironments. The need to run a railroadon a day-to- day basis fiurthercomplicaues the process. This paper is pamtof a seics publishedby CFS, and is a sequd to an calier volumcon the privatization and zer staue reorgaization ofJapaneseNational Raihways.The seriesis intendedto communicarethe experienceand insighs of priatizntion practitionersto their colieaguesaround the world, and this most rent volumneaccomplished these objectives admirably. On behalfofCFS I would liketo thank di authors for their valuablecontribution to the art and scienmcof priraciation.

Inder Sud, Dircaor, Cofinancingand Finandal AdvisorvServices Acknowledgements

Theauthorsexpressspecialthan to Mes TakayukiNayakadAkraSatoofdrejapan Development Bank who fonneda teamwith authorsand assistedin the completio:-ofthis study.The authorswouldalso like tO hk two Proessors of HamrardUniversity, Dr. John R. Meyerand Dr. JoseA. G6merz-Ibafiezfor their constuctive advice.This paper was initited as a component of the World Bank Study entitled "Compartive Analysisof Railrad Privaization.' managed by Mr. Ron Kopicic of Co-financingand FmancialAdvisory Services, thc World Bank,but also has been completedas an independentstudy and constiotes the sequelto the lreadypublished Wodd Bank DiscussionPaper, 'JapaneseNational Railways PrivadzationStudy, Japanese Expeice and Lessonsfor DevelopingCounties (No.172, 1992). The authors expresstheir appreatn for the insihtfisueion of Messrs.Louis Thompson, Kvin Young and DhrmvaSahaioftheWorld Bank,HiromichiToya and Kunio Kishimoo ofthe MinistryofTransport in Japan, Shuhitro Yamanouchiof JR East, an YasushiTanahashi ofJR Freghr. The authors also xpressappreciation for the informationalsupport of Messr Akio Koike of JR Hokkaido, Knlo Aold and Toshini N*gswa of JR East, Ksumi Myawa, MasayaSakta and Mahiko Emumaof JR Cntl Makol Shibata,Masshi Nonaka and KojiNakura of JR West, Hachiya Hirono and MicdinobuSogava ofJR Shikok, Masahir Iimamuraand KenjiMon ofJR Kyushu,Yasuo K cluiandTahshiAbeofJR Frgh MinoruSuzuki Masaito MoriandYujiSegawaofJNRSetdement Corporation,YodLiali Komasu of the KinidNippon Railtay, YasuhiroKao of the NagoyaRailroad, and Fumihiro yaA, Hiddei Napo and ShigeyoshiFujiar of the Japan Devlopment Bank. The authors finaly would like tOCxpress their gratitudefor the generoussupport of MissAyano Takei and other stiff fiom the Intenational Coopertion Deparment of the Japan DevelopmentBank. IntroductoryNote

This report followsan carlicrreport by the same authors investigatingthe resultsof the 'privatization' ofJapanescNational railways UNR). It takesalongerrangeviewoftheoutcome: what happened,why,what succeded and what did not? Observcrsofthc railwayrestruacuringscene are fortunate to havethse reports becausethey finallypermit an answerto thesequestions. It is no longer necessaryto sayr 'Results look promising,but it is premature to daim succes" Now, for some questions,there arc answers. Byanyrcalisticmeasure the processwasasuccess.Traffichasgrownstronglyonallpartsofthenewsystem becausea profit orientationhas made indiffcrcnceto customrs a luxurythe new companicsURs) cannot afford. With revenucsup, hugcdeficits havc becnconvcrted into solidprofits through increasedattention to cost control. A bloated laborforcc has becn reducedand labor productivityhas grown rapidly. One of the railwaysUR East) is wcll along towardtre privatization,and credibleplans to scll shres in two more (JRWestandJRCentral) havebeenestablished.Mechanismshavc becnestablished topermitpublic-private partnershipsto financethe heavyburden of newcapacity expansion. If the Governmenthad not chosento intervenein the sale of railwayassets(because the real estatemarketwas overheated)and if the stockofthe three HonshuJR's had beensoldbeforethe "bubblec economyhad burst (ra thanwaitingfor thccurrent marketwhich is too weak),all of the originalJNR debt woult alsohave bcen repaid instead of fallingmore tan $ 100billion short. All inall, it isdoubtfiulifeven theoriginalproponentsofprivatiza0onactuaLlyhoped forrcsultsasfivorableasactuallyhappened. Certainlyther arefewgovenmmentsesewhcrein theworld that took on such a largc problem and made so much progress. The reasonsfor success,while alwayshard to implement,are simplcin concept. Definc objctives dcarly and credibly,and build a consensusfor reachingthe.r Whcre an adequatemarket easts, adapt tkh solution to the demands of that market ratder than remaining in a government nodc of operation: in particular, mak maximum use of the pressuresof compctitionand the techniquesof rhe private sector. Finally,identify winners and compensatclosers. But the reportalso highlightsa seriesof importnt issuesthat remainto be resolved. These are * TvffRquLzion. When and how will thc Govemrnencallow the JR's to have more rate flesibility so ticy and the large privaterailways can expecttO earn adequateprofits? * The threeHonshu Raihmay 'WiUprivatization be completed? How (unlike the privateraiwas which are predominantlyurban passengeroriented) will the Honshu JR's continue to managea mixtureof suburban and interciEypassenger seruvices which might be better handled by sepaate entities,or even transfered in part to the priva raiways or localgovenments? * Th thre d raikvys. The revenuestabilization fund mednism is unstable (becase offluc- tuations in inteest rates)and is a barrir to privatization:should it be replacedby a contract for public serviceswhich would be pxedictableand could easilybc tmnsferredto a privateoperator? Should the local govermeCntsof the threeislands be askedto make a contribution to the contact paymentsneeded? * The Fnght Railway. Is there a rationalcost-sharing formula betwe freight and passengerwhich would not price freight off most main lines? Is dtere any hope that freight could affbrdto pay for capacityexpansion - and can the countryaffort to pay for more highwaycapacity expnsion? Is there a better balancebewveen private and public rolesin rail fieight sevices, and (givean agreed method for line operacingcost allocation)would it not be possible,even imperaive, to privaize JR Freighi through a limited auctionor evena compettve, privateplacement? There is much the Bankand itsborrowers can learn from thejapaneseexperience and, giventhe issues which still remain,there will be good reasontO kp a closeeye on developments.

LouisS. TIompson Raia Adviser The Wodd Bank ExecutveSummary

The fis half of the present centurywas some- 1970s.In particular,the deficitsgencrated byJNR, thing of a goldenage for railwaysin many coun- the Foodstuff Control SpeciadAccount, and the tries. In Japan the first railwaywas built between National Health Insurance System posed a real Tokyo and Yokohamain 1872 by the Govern- problemforlgovernmentfinance.Expendituresfor ment. New line construction folowed in various thesethreeitems aloncaccountedfornearly lOYoof areasand by the rurn of the centuzythe mileageof the tota budget. In March 1981, the Provisional privately-runrailway lines exceededthat of Gov- Committee on AdministrativeReform was ap- cmnmcntowned lines. Immediatelyafter the Russo- pointedunder theCabinet's jurisdiction andcharged JapaneseWar (1904-1905),the JapaneseMilitary with divisingdrastic measures necessary to achieve becameconvinced of the raihvay'sstrategic impor- fiscal reform without raising taxes. In 1982 the tance and lobbied for nationalization.The Rail- ProvisionalCommittee proposed the privatization waysNationalzation Lawwasenacted in 1906xnd ofJapan'sthree largestpublic corporations-JNR, cxistinglines, both private and pub)ic, were inte- Japan MonopolyCorporation (tobbaco and salt), grated into a nationwidenctwork under Gove-n- andNipponTelcphoncandTctegraphPublicCor- ment controLThereafter, massrvc construction of poration.The commiteealso suggested that private new knes continuedthughout the country.The initiativebe employed in the constructionof social decelopmentof the railwaynetwork provided the inrucre, which in the past had been hieved basis for a more mobile societyand workforce.It primaily through public works. alsofacilitated the spreadof a more homogeneous Of all the ProvisionalCommittee's tasks, by far lifistyle, rapidly diminatingtraditional regional- the largestin scaleand most crtical was the priva- ism. tizationofjNR.Japan'slargestpubliccorporation. In the larterhalf of this cenruy, cspecallyover In 1980,JNR cmployedncarly 414,000 workers, the past two decades,the railwaysector has experi- out ofwhich about onc halfwereconsidered to be enced coughcompetition from diffcrcnt,increas- redundant,andthecoporation'shugeannuallosses ingly sophisticated modes of uansportation. and accumulateddebts were increasing yearly. On Stat-operated railway systems almost universally the other hand, JNR maintained superb cnginecer- began to losemoncy and to requiresubsidies. The ing and opratirng ehniqucs ant the transition situationcontinued to deteriorateatan acclerated had to be handledin awaythat did not diminishits pace. Thus, restructuringof railwaysystems, in- tenial strength. Thc Provisional Committee duding the abolitionofservircs, has becomea high recommended the cscablishingof a Superisory priority issueworldwidec The Japancsecase pre- CommitteeforJNR Reconstruction,to formulate sents an interestingexample of restructringrail- and implement concrete policies for the ways,not onlyin its sizeand methodologybut also corporation'sreorganization. In 1987, in line with initsphilosophythatdcrailwaysmustcontinueto the SupervisoryCommittee's proposal, JNR was play an important role in the future, that the brokenup into SiXL .gionplly-based railw2ypassen- burden to dtc nation must be minimzed, and that gcr companiesand a sventh company to handle this type of retrucuring should not occur again. frcighttransporcmdionfortheentirecountry(These Sevenyears afier the successfilresuuriing, this concernsar referred to as the JRs in this paper). paper examinesthe rnaining and newlyarsing Abouttwo thirdsofJNR's debt,wiichhad reached issues,and makesan economicanalysis of raiLway $337 billionby the end of 1986,was reassignedto privatization. the JNR SerdementCorporation along with any Brkef DCSCT~puiOUeflapanese Nauonal ~ surplus real estate of JNR and the shares of the newlycre ated JRs. Ile eliminationof the redun- dant workcorc was carriedout smoothly,aided in Asinmanyaodercountrieswhichhaveachieved part by economic growth and by the vey low or are considern privatizationof the railwaysec- unemploynt rate of the latter 1980s. tor, printization of Japanese National Railways The performanceofJRs sincerestructi has MNR)was tniggeredby the financialcrisis of the beenremarkmble,especiallyuntil 1990. Theannual Japanese Goverment in the latter half of the incease in passengersand cargohas been 5% and u Japanwelt National RAilwsPrivstiAtion Stdy I 10% respectvely.The number of employeeswas As a result, the proceedsfrom the saleof assets drasdcillyreduced to 191,000 by 1990 and trans- have not coveredintcrest payments, which totaled portaon efficiencyhas improved remarkably.In $56 bilionbetweenFYI987and 1992. In faa, the 1991, cheJRsasawholepaid $4.0 billionin taxand liabilitiesof the JNR SetdtemcntCorporation had contributionsto the government,even though rail increasedto $242 billionby the end of the FYI992. fareshad not been increasedsince 1987. JRs are In 1991 land pricespeaked out and showedsharp mulingefiorts to respondbetter to localneeds and declinesubsequendy. According to the figurcsrc- their corporateimage has improveddramatically. leased by the JNR Stdtement Corporation, the InSeptember1993, 62.5% ofJR Ease'sshareswere valuc of this land bad on the posted price as of offbred to the public. The renaining shares of JR April 1993,was roughly$76 billion.In the lightof East,aswell asshares inJR CentralandJRWesc, are this, the Government has lifted the ban on com- scheduledto be sold during fiscal 1995 and later. pCitive bidding, but the oudook of land prices is The restructuringand privatizaion ofJNR can be still bleak and salesof the Corporation's land are judgedtohavebeensuccessfil,althoughthegrowh sti slow in spite of variousmeasures to promote of passengersand profitabilitybegan to stagnatc land sales.The repaymentof debt, the major re- from 1992owingtotheprolongedeconomicreces- mainingissuc to be rcsolved,must bc carried out sion, and thcre cxist remaining or newlyarising rapidlythroughthesalcofsharesandland, orbythe issues. xmition of profits from the developmentof Dehu dftJN Seaement Corporation ral este. One unique chamactristicof thbJNR privatiza- Sa ofJRs Shar tion is theway inwhichjNR's debtswerc allocated It was planned that as soon as the JRs met the to variousbodies JNR debts totaled$337 billion, listing rquirements oftheTokyo StockExchangc, induding fumtreliabilities. The JNR Scetlement theirsbareswouldbelisredandsoldandprivatiza- Corporadioninhiedatotalof$232billion,outof tionwouldbecompletedThrecoftheJRs (JREast, which $26 billionwas scauredas a ftumr leasefee JR CentralandJR West ) alreadymet the require- in respectof Shinkansen(bullet trains). According mcnc by FY1991and salesof their shares were to the initialplan, the Corpoation wasexcxted to planncd to beginin FY1992.However, though the repay the remaining$206 bilion throughthe sales basicprinciple of theJNR SupervisoryCommittee of surplus assets ($70 billion) and newlycreated that all the sharesof JRs be listedand sold tO the JRs' shares($11 billion)with the balanc of $125 public has not changed, the time schedule was billiion to bc bom by the GovnmmenL delayedbeaus ofanxietiesthatamassiverelcaseof Soonafter the formationofthcJNRSettdement new stocksmight fiher depressthe alreadyweak Corporation in 1987, the price of land in Japan stck market. begn to soar and greatlyapprecited during 1987 In September,1993, 62.5% ofJR Eases Shares and 1988. Had all the land been sold without were offered to the public, as noted above. The constrint by the peak of the emnomicboom in r.mi 1.5 miion unsoldJREastsharcs,aswell 1990, the proceedswouldbave met the the liabilities as thoseofthe JR Central(2.24 millionshares) and of the Corporation.Howcver, land saleshave been theJRWest(4millionsharcs),wcrescheduledtobe cared out at a much sowr pac than itRy soldfromfiscal 1994 onwards.In December1993, phanned,owing to the Government'sdecision in the Vice Minister for Transportation announced October, 1987 to prohibit compeitivebidding for that only one ofJR West andJR Gental would be dte readesate of the Coxporation.The Govem- soldoffinfiscalyear1994, indicatingtheabandon- mentmadeitsdecisionarthecevnseofandagainst ment of the intention of the Ministry to list both the objectves of the Corporation, becauseit was companieson the stockcxchange simultaneously. thought tat such biddingmight pushgeneal land In Junc 1994 the Asset Disposal Council of the prices up further. Thrforc the saIcof real estate JNR SetdementCorporation conduded that only has been limited. The sale of JR shares has becn JR West should be listed in Fiscal 1994. Thlis ddayed also,owing to the depressedstock markeL decisionwas made in lightof the postponementof kwas only in Sptember, 1993 that 62.5% ofJR thesaleandlisdngofJapanTobaccoInc., (anodtr East'sshares were offeredto the public for dth first Govemment-ownedcompany) fiom fiscal1993 to aime, fiscal 1994. In October 1994, one third ofJapan Japmae Naional Raiwaysr Privasian Smd H w Tobacco's shareswere offerd to the biddersand encedexcess capacity in the pastfew years (though public, but a substantialportion of tese shares demand was above the capacityof existingfeets remained unsoldbecause the sellingprice was too during the period of the "bubbl" economy).The high. On November 15, 1994 the Government situationworsnedin 1993owingto theconfision decided the sales of JR West's shares should be of cargoescaused by unusuallyfrequent typhoons postponed until Fiscal 1995. The Goverfnment in the southern part of Japan and to the drstc consideredthat the absorptionability of the stock reductionof ricecrop transportationcaused by the marketfor privatzedcompanies was not sufficient. abnornally cold summer. Owing to this decision, the completion of sale In line with operatingvolumes, the revenueof proceduresfor the sharesof the three Honshu JRs the JRs also increasedremarkably up to FY 1991 has been delayedstill furither. and then stagnated, because fiLrcshad nor been As fras thesaleofstocksofthe three IslandJRs adjusted since 1987. On the othe hand, certin is concemed, in addition to the weaknessof the cost componentsof theJRs continuedto increase, companies'profit baseowing to fallingyields on resultningthedcdineofbothoperatingprofitand the ManagementStabilizing Fund, the argument retained eamings for the most of the compnies still remains unresolvedover whether or not the during the past two year In addition,in the caseof listingofa companythat is not derivingoperatung the three HonshujRs, the purchaseof Shinkansen pro fits from its main businessis appropriate. Be- lines in 1991 has had a significantnegative impact causeof theseproblems, the timingof the disposal on theirprofit. For the threeisndjRs, recentlow ofthe three IslandJRs'stock is not dear at present. intres r ates have decrcasedthe yieldon the Man- As regardsJR Fright, in addition to the need for agementStabilizing Fund. ntaining a cermin levelof profits in order tO Thus, the environmentaffeaing eachJR has aieve lsting on the exchange,many ocher issues incrasinglydivergInFY1993thetheeHonshi remainunsotved, such as costsharing with passen- JRs still adcievedadequate profits, thoughin the ger JRs, the burden of new investments,competi- case of the three island JRs, their profit levelhas uionwithntrucks, rricuons on expnding:Yusiness declined almost to breakcven leveLJR Freight and othes This will make stock disposaldfficulct rcorded its first deficit in FY 1993.Ncvertehess, in the ncar f&turc. efforts to respond better to local meedsand an b&'RecntPer Jimprovedperformance in gel have been ob- -femaxe servedineachJR. This isdcearlythepositive sideof In the first four yearsup to 1990,the respeaive privation. annualincreasein tpassengersand cUgDhad been 5% and 10% - madring a dcar contast wih the Is of Li C decliningtrend up to 1986. In 1991 the growth of Railwy business, regardlessof owneship, has passengersstill showed an increase of 4.9%. In somc public aspectby naturr. In the days ofJNR, 1992, hower,thevrahemofpassengergrowthslowed tbis public aspect was emphasized.Since rhe re- down drasticallywith thEcend of 'bubbl" eco- stucuring in 1987, however,the pendulum has nomic growdt in Japan. JR Central sufferedmost swung to the opposite cxtrme and priva;izedJRs among the three HonshupassengerJRs, as it rdies are reluctant to undertake new linc construction. hcavilyon businesspassengers using bullet trins. This is one imporant area where the conflict of As a result, the companyregistered its first negative public interestsis grwing betweenthose who bear growth in passengers.In chat year the growth of the costof, and thosewho receivebenefit from, the passengertraffic for the passsengerJRs as a whole new Shinklnsen lines, metropolitancommuting wasonly 1.0%.The prolongedrecession continued lines and intcercitylines. in 1993 and hree passengerJRs (JR Hokkaito, JR In the case of the Shinkansennetwork, a devel- CcntralandJRKyushu) ceriencedslightdedines opment plan was formulatedin 1973to add to the in passengergrowth. The passengergrowth for existingfour lins. This plan was tempoarily put passengerJRsas a wholewas a mere 0.2%. onholdin 1982duringtheprocessofJNRrestruc- The Japan Freight RailwayCompany, which muing.Since then, demands bave been raised re- registcrednegative growth for three consecutive peatedlybylocalinterestsandbyamajoriyofDiet years,was most severelyhit, as it had to compete membersfor the rapid implementationofthe pan against commercialvehicles which have experi- In 1989 the Government prioritizedeight projects x Japanse NaiondAl hval Primdaion Std and constructionstarted. The priciple of burden necessaryfor any railwayfreight capacity increase. sharing for JRs is that; (a) the JR concernedbears Acordingly,the lcase rae was increased,whch had 50% of the constructioncost and (b) lease pay- amajornegativeimpactontheprofitofthecDmpany. mentsby the JR concened are commmsuratewih JR Freightalso has to continuouslyrenew rolling the benefitwhich the newline concernedproduces stock.At the sametime, competition with tics has For otherremaining projecs, which are estimatedto been intensifiedin the past few yer owingto the cDstsome $50 billion,neither constructoio period, prolongedrcesionandvab ityoftherailwayfreighr nor priority,nor the find sourcewere deermninedL has been calledinto questionai In the case of metropolitan cornmutir4 lines, In the face of externaldiseconomies and long- private railwaycompanies in Japan tradidiocally term labor shoragcs affecting road fright, the dcvlopednewlinesandinternalizedextemalecono- Governmenthas examinedmeasures for a modal miesofsurroundingrealestateapprciation. Nowa- shift of freight traffic from road to railway, or days, extemaleconomies are dividedamong many maine, and for the acievmnt of multi-modal concernsand it is impossiblefor railwaycompanies trasportation. In 1991, the Govrnment decided to recoverconstruction costs. In addition,negotia- to promote a modal shift to rail through interesc tions to furthert evelopthe nework havc become subsidiesand tax reductions on certain typesof increasingycomplicated and involvemany inter- investmentbyJR Frcight.Further steps,however, csts. For thisreason, the conceptof "Public-Private wouldraisequestions about thelong-termviability Partnership"hasbecome common in constructing of rail firight in comparison with other traffic new metmpolitanlines, where a newlyestablished modes,especiallywith :rucks. The issueis compli- corporation Ioindy owned by local governments catedbyallegedcrosssubsidizationekoistingboth in and private concernsconstrus and operates the train and rudc transportaion. Furdte complica- constructedline.In manycases,railwaycompanies tions arise if exnalities such as congestionand willnot agreecvento bemme operators,suspecting pollutionare taken into account. tha fiuure financialobligation migt failon them. Using current maret pricesand without con- Thus, actualprojectsarc limited, although the need sideringexternaliis, an indicativeeconomiccom- exists,owing to the rapid spread of metropolitan parisonis attempted in this studybetween railway areas.As for intercity lines, largc needs therc scill androad feightinthccaseoflong-distanccfreigh, Cxistfor the speedingup and capacityexpansion of wher rlws may hvc relativeadvantges. The lies. Here, "Public-PrivatePartnership" is also rcsultshows that the costofrailway is a litde lower sought,asJRsusuallyaskforfinancialsupportfrom than that ofroadfieight. In odhcrwords,rail freight and participationof localgoveunments. can compec with trucks in the long distancemar- krCon the basis of current costs. Thus, the recent mpredicamntof JR Fright is seen to aris from a The fieightsectorhad accounted forisubsrzntial lakofLmaketingabilityorbargainingpowerwith portion ofJNR's total operationaldeficit through- the final dients, becausecontainer freight by rail out the 1970s and up to 1986. By the time of must use trucks anywayat both ends.JR Frcight restructuringin 1987, railwayfreight was playing alsohas another constraintin that it cannot expand only a marginalrole and the perceptiongrew that operationsofitsownwillwhenaconfliccariseswith it was difficult,if not impossible,for this sector to passengertwains for the usageof rail. be profiaolc. For this reason,in 1987,JR Freight C . o btwJRVAwdPri assumedresponsibiltyonlyforroingstoc inorder to minimize pital costs. It was decded that Passen- gerJRssbould own the infraucure and payments Few people disagreewith the successof JNR by JR Freightfor rai lcsc were deemedto be an restructuring,comparing the efficencyofJRs and aavoidablecost", which is basically the CoSt ofincre- JNRI Howevcr,the questionremains as tO to what mentalecrity and maintenancecaused by fiight cxtentJRs are operatingefficiendy in comparison trafic. Fromabout the same time when JR Freight with priva ailway companies(PRs). In Japan, begant opecza, traffic congestionon the road and thercisalargenumber ofprivate raiWacompanies the shorageoftruk driversbecamc sious. This led in addition to JRs. PRs havehistorically rcorded tO a rapid increasein frcightrailway traffic and in stable profit levds (despiteinxibk firc regula- 1992, JR Friot agreed to bear the capital cost tion), thians to complementaryside busnesse Jlapse Nadnl RaiJwys Privanution Study f xi such as realestate development. Among them, the panie&AtthesametimethelocalorienrationofJRs largestfiflen privatePRs (LargePRs) in the three has also introduceddirect competitionwith PRs, mceropolimaareas (Tokyo, Osaka and Nagoya) especiallyin paallel commnutin-lines in major haveplaymdanimporantmleinuansportingpeople cties where a JR and a PR provide competitive to work and schooL(he achievmnt ofmanage- semes. On these lines, competitionhas contrib- menton aparwith that ofLargePRs was oneofthe uted tO improvedservice quality in terms of train guidelinesfor the JNR prnaization). freqncy and load factor,andhas help-d to hold A comparison of the thdre Honshu JRs and down the pressur for fare incrases. It is wvorth large PRs using simple average figuresin 1991 examining to what cxtentthis intramodalcompe- shows that the former bad almost caught up in tition, or benchmarkcompetition. has been inten- efficincy in certain key respects.In terms of car- sified,or has contributed to the growth of market kan, the JRs wereable to decreasetheir numberof share fbr JRs. employes per unit to a level similarto that of the Regarding the frequencyat peak hours on se- largePRs. In terms of passenger-kmper employee, lected lines, the study shows that JRs more than whilestillbelow that ofthe LargePRs, thejRs were doubledthe frequencybetween 1975 and 1992 on able to decreasethe differencefrom 46% in 1987to lines where competidon is strong, whereas the 16% by 1991. In tcrms of cost per car-km, JRs' increasewasonly 19% for lineswhere competition costsare stil higher than those of the Large PRs, isweak.Further,thestudyselectsdthcommung howvcer.Regarding fares, the JRs have not raised areaswhere aJR and a PR are opating in parllel them fiom 1987,while the LargePRs have maised in order to assesshow benmnark competitionhas theirs by roughly20% firom1987. As a remlt, thc influencedmakt share, or psengers, of JRs. In differcncein farelevelshas been dedining.In some theselines JRs have kept faresrlatively low, hae areas,theJRs are carying passengersat lowerfares improvedcongestion problcms by upgading cars, rhianthe LargePRs. and havc increaed tnain frequency. Thc result Labor productivity (passcger-km etc.) of all showsthat JRs arc acquiringmorc passngers than assengerJRs was genrally cnhancedby a factr before,in somecases doubling their marketsharc. about thru timcs between 1981 and 1991. The Therelativeincreaseintrainfequcncyseemstobe improvemnt by job type is also examinedin this the key tO success,sLiCe tberc does nor existany study. The analysisindicates that JRs improved pricewar betweenJRs and PRs.Thus, benchmark productivityin station and car maintenanceactiv- compettion has improveddte levelof servicequal- ties and lessso in truc maintenanceand adminis- ity in the commutermarket and the benefitsof the trative activities.JRs' productivity was stil about improvementhave been passedon to commutcrs. 20% lower than Large PRs, but in the case of m ofjR Fright operatorsand conductorsit was about thersame. Simpleaverage comparisons such as thesccould In Japan, passngerJRs own railwayinfiastruc- bemislading,however,sincetheJRsandtheLarg mreandJR Freightpays userlease fes basedon the PRs have diflirent operating environments.For avoidablecost. In this siuation, one important this reason, the study also cmployed economic questionintheeconomicsenseiwhthruse-rlcase analysis using regrcssiontechniques which rake fecscover the relevantcost of tackuse, or marginal into accountdiflerences in nctworksand scrvics cost, from the stndpoint of efficiency.In this The resultssuggest that passengerJRs still retain study, the cost function of passengerand firight workersperunikofoutput20% wt30%inexcess of sevices is examinedand the marginalcost of JR those e mployedby LaugePRs in most activities Freightis estimated,using an econometricmodel. except opertors and conductors. Howevcr, this The study shows that the margia cOst of JR differencecould be explained by morc usage of Freight is much higher than the user lease fce, contracting-outby Large PRs, or greater conges- indicating that JR Freight is cross-subsidizedby tion in LargePRb' cars at the cost of passengers. passengerJRs. However, the extcnt of subsidiza- Be.cknrk.wedonamong and vate rRstion that can be inferred from the rcsult must be .e...mark Cempedden amongJAr and Privae treated as ndicaive only, since data used in the Iwav time-scricsanalysis represent the period of drastic The regional break-up of passeger JRs has JNR restructuringand may nor reflect the actual introducedindirect competitionamong the com- relationof freightvolume and incrementa cost. xiiuapawe Nationd Aivdays Aris on Study H T1hecurrent aoidabl-costmethod istied to the that the benefit of the improvement has been particular railwaypolicy to maintain rail fieight. passedonto passengers.Freight rail transportation On the other hand. uucks are alsocross-subsidized is still cross-subsidizedby passengerrail, but this by passengercars. Furthe, if extemal costs are issue must be consideredin the contex; of more associatedwith road transportation,therearesound comprehensiveinter-modalcomprison,sincetrucks economicreasons for railwaysto deviatefiom mar- are alsocross-subsidized by passengercars in terms ginal-costpricing, Indicativc economic comparison of lowhighway toils and fuel taxes. betweenrailandtruckfrcightintheprceding secion The restruturing of railwaysystems, induding showstdat rail feght can bc compedtvc cost-wise the abolition of services,is a phenomenonwhich over long distances,indicauing the validity of the has occured in many partsof the world during the current policyfor rail freighL past twenty years. In Japan, this has been caused La fJNR Priatio mainlyby the rapid developmentof automobiles, vwhchoverwhelmedrailways in manyareas,and by This study focusesprimarily upon the various the massivemigration of populationto metropoli- issuesarisingsevenyearsafter the initialrestructur- tan areas (especiallyto 'rokyo), which has made ing of JNR in 1987 and upon economic implica- many locallines unprofitable.Similar phenomena tions ofits privatization.For thosecountnieswhich have been observedworldwide. are consieng rsructrg now, the reallyim- In the long run, however,the viabilityof the porrant lessonsderived from the initial restrucr- railway sector in comparison with automobiles ing in Japan, which are eamined in detail in a must be analysed with care, when external preerdingsudyC"JapaneseNationalRailwaysPrna- diseconomiesof automobiles,the toughest com- nation Study-The Experienceofjapan and LIs- peitor of railways,are taken into accountsuch as sons for Dcveloping Countries", World Bank congestion,pollution, destuction of nature and DiscissionPapes 172 ), are still valid.These in- glo warming.At the samedime, over-ncentra- dlutc impartal planning, prioritization of issues, tion ofpopulation in Tokyohas causedintolerable creation of incentivesfor effEcientmanagement, problemsfor both metropolian and rural inhabir- and compensationfor los However,even with ants and there existsincreasingly a strong demand th inital restructuring succesfly compklted, for more balanced developmentof die Japanese many counmes may eventally fiLacesimilar issues archipelago.Here= railways can play an important to tose which are disimssedin this study. role, consisent with the commercialprinciples of Our economicanalysis shows tiat the produc- JRs.ThercstrucuringandprivatizarionofJNRhas tvity of JRs has been gready improved by dte been based on the phiosophy that railwaysmust privatzation,which has madc the nation the big- playan important long-termrolc. In each county, gest beneficiary.It also showsthat rhe quality of decisionsmade today could have far-reachingef- serviceshas been improvedby competition and fects upon later generations Tableof Conte

Chaliter ...... Pago Foreword...... uu Adknowledgements...... h InrducoryNote r...... v ExecutiveSummary ...... i I.Privadzation ofJapanew Nationd Raways ...... 1 PHiatization inJa pS6Wand itsBe1k9g ad...... id...... I CH& withintJNRanditrCasa ...... 2 PhrivizioniPerossofIf ...... 4 Perr.rmnancef..wihngjNRR a¶uoxng ...... 5 RemainiamgIsa ...... 6

H.Rcent DeveopmensAfedangtheRemainingIsues ...... 8 Lsauesofthejapiae so Naiw Senap iwaCo seio... .. p....n. . 8 DeC.ofJNRSC. 8 AefNofR...SC. 11 LandsalesofJNRsc . 13 Poeles to ExseReguleow Conxemixgthe Dkp.s6LdfLond . 13 Measresfor LandDiposa withoutDisdosixgLaxdPkes ... 14 JrssuCsCon..ingS.ks .f..' as.w. . 15 Peficieso he S of ,aShar.15 *fs Iisial Sals ofJR EastoSha .s . 16 77mrSthedlefor Fse Sbwr Sak.. 18 km Safor NoHmHs....oh...... 18 I, JR:? Recent Performane and Newly Ariing I .s ...... 20 JRlsRec xtPermaxce ...... 21 Recnt Deueupze off/s' featiox.21 ProJfiabi&,vzdMAfajorConce ofsj ...... 21 ProibityorDivvi*dBusixs ...... 23 capitdE.pexnditwswandFdexdnaFmlin xgofjr ...... 24 FinnzcidBurdn on Gonrnmexnt...... 26 Eff&n*ondidprobfi= ..ofBs ...... 26 Thre Honshu ...... 26 Thee IsnvdjRs ...... 28 JR Fright ...... 32 New Li Contrucion and ConflictsofPbk Int t ...... 35 GrowingConflics oferst ...... 35 Sinkanseni Networks ...... 38 Mebv&troitanCommuting Lies...... 41 Inercity Lins ...... 43 sssesofiWight Tramporttimn ...... 44 ArgumentofMoad Shift ...... 44 Avoidabl CostMethod ...... 45 GrowingConJtiofIntfrts ...... 46 Competitionwith Trucksand xndIe Ecnmic Comparisoxn..47 Begantioxby Govrmnext .47 Contwd IV. Market Strclture of Transportationand ComparisonofJRs and PrivateRalways ...... 52 AfarAetStucture ofRailroadSector in Japan...... 52 Radrod SysteminJan Jap ...... ,52 RadroadTraffic Charatecs ...... 52 PazengerTraffic ...... 56 rAhgbtTrafic ...... 56 ComparisonofPerfmuawe betweenJbsand PrivateRailways ...... 8 LasgePrivate Railway ...... 60 Comparisonbetween Hoshbu IRs and LargeP ...... 65 V.EconomicAnalysis of Productivityand Performanceof PrivatizedJapanese Rlways ... 74 Pr.dwviity Comparisonbetwen JI nd Privae ailways ...... 74 AnalyticalComparison between Ri and PrivateRailways ...... 74 LaborProductity Growth...... 74 LaborProductivity by Acivity ...... 75 Phdctivity Diqeoncs amongJRs...... 78 EcononicAnaysisofLabor roductivity...... 7p enchmar CompetitionamongJRs and betweenJRs adPr;P i ...... 81 The CoxneptaFramework ofBenchmarkCompetion ...... 81 OverallAssessmentofBmnchmark Competiin ...... 82 T fetofBh markCompition onJRSMarketShas ...... 84 AppexdiA.A Sivmk IntegrutedAnalysisof Bewhark e ti ...... 88 EconomicAnas ofJR Freight...... 92 Changesin Costand ServiceOutput ofR Frez.t...... F ...... 92 Estmation of Ct Funion ...... 4 Estimatn of CooFwunaonofJ&r ...... 95 L Privatizationof JapaneseNational Railways

1.1 Asinmanyothercountrieswhichhadto readhed39.6% of its tol budg This levd was consider,or are considering,privatization of the extremelyhigh when comparedwith that ofother railwaysector, privization of JapaneseNational leadingWesten counuics, whose dependenc ra- Railwayswas triggred bythe financialcrisis ofthe tios in 1979 ranged from 5.6% (United States)tO Japanese Government in the latter half of the 14.2% (WestGemiany). 1970sLSnowbaling debts and deficitsofJNR and 13 Three factors contributeddirecdy to this the resukingfinancial burden on Goverment had crisis:first, prices andwages injapan had increased reachedan enormousand unacceptabklleve by dmntically in the wake of the oil crisisof 1973, 1980.The rdativededineoftherailwaysecmorwas resuling in gradly inacased expediur for so- in a sense inevitabledue to the devclopmentof cial security, educaion, and other areas where dieet, in,aeasinglysophisdcardmodesofians- Japanwas just beenning to achie pairywith the portation.JNR could not copepropely with this advancedcDunties of the West, second,in 1974, change . which acclerated the deterioration of Japan experienced gatvc economicgowth for JNRs financialhealth Afierthe failureof recon- the firsttime since WorldWar I. After 1976,in an suuction plans byJNR managemen and frequent e&irttO achicvc a more stablelevd of groth, the strikes by JNR labor unions in the 1970s, there governmentgrearyincreasedthenumberofpublic existeda widesprcadperception by 1980 that the worksprojects undeaken, third, the slowdownin restucturing was essential.On the other hand, economicgrowthledto areduction in =revernues, JNRmanained superbengineeringandoperating whicAin turn led to increasedgovemment expen- techniques,as was conspicuouslyshown in the dirures.In order to cover these expenditures,the bullet train nctworkespecially. The transitionhad governmenthad to issue more bonds. As a result, to be handled in away that did not diminishJ's the govenment was confionted with an incrased technicalstrNgdL burden inmeeing principaland interestpayments Prwaiuation in Japan in the 1980s and its onthebonds. Backgrod.1.4 In addition to the economic and social Background changeswhichoccurred afitr 1973, another factor 1.2 Bythelatterhalfofthe1970s, government contibuting to the fiscalcrisis was the continua- financein Japan had reacheda crisis. In order to tion of spendingpracces which had been greatly respondt this crisis,in 1975 the government ecpanded during the period of high economic beganissuingspedalbonds(so-led deficitbonds) growt The country's financialstructure was in- to hdp cover administative cxpenditrcs. These flexibleand consequendy it was very difficult to bondswere in addition to the constructionbonds reduceexpenditurs. In paricular, thedeficitsgen- the governmentwasalready issuing to coverinvest- ematedby the Japanese National Railways(JNR), ment expenditures.By 1979, bonds had been is- the FoodscuffControl SpecialAccount, and the sued fr amtal ofUS $521billion (C57.3 trillion)' I H US$is abbrviatd as S.An ccangemm of and the goverments dependenceon them had 511110sse. 2 Japanese National RailwaysPrmiizution Study II National Health Insurnce System posed a real administativc approval and governmentfinding problem for govermnent finances. Expenditurcs were reviewed,(b) the three public corporations for thcse threc items alone accounted for nearly mentioned above were privaEized,and (c) large 10o of the total budget. projects, such as the Tokyo Trans-BayHighway 1.5 In rcsponscto thcscproblems, the Minis- and KansaiInternational Airport (in Osaka),werc try of Financepresented a fiscalreform plan to the implementedon the basis of privateinitiaive. Dict in 1979, the stated goal of which was to 1.9 Of all the ProvisionalCommittee's tasks, 'reduce the issuanceof specialgovernment bonds by fiLrthe largestin scale and most criticl was the to zeroby 1984."In support ofthis plan, the Ohira privatizationof JNR,Japan's lagest public corpo- Cabinet, composed of Liberal DemocraticParty ration. In 1980, JNR employednearly 414,000 (LDP) members,tried to introducea gencralcon- workersand its huge annual losscswere increasing sumption tx. But public opinion was strongly yearly.TheProvisional Committce conduded that againsrsucha tamComplaintswerevoiced tha tte rcstoringJNR's financial healh wasimpossibk as Government had not made sufficientefforts to longas it wasoperated as a centralized,nationwide streamlinc its own operations, particularlywhen public corporation.The Po-isional Committee comparedto the cflbrt that the privatesector had rcommendestablishmenrofaSupervisoryCoM- mnadein ovcrcomingthe oil shock The public felt mitcee for JNR Reconstructionto formulateand that there was considerableadntinisurative waste implent concrctepolicies for the corporation's andthat educinggovernmentexpendituresshould rcogaization. In 1987, JNR was broken up into bemadeapriority.Asaresult, the LDPwassuffered six regionallybasedrailroad passengercompanies serioussetbacks at the polls. and a seventhcompany to handlefreight transpor- 1.6 Thus, public opinion madc it difficultto tationfor the entirecounty (rkerred to as the JRs rebuildgovmnment finances through largetax in- in this paper). The major portion of JNRs debt crcases.In 1980, however,the LDPwon the gen- whichhadrachedS337biionbytheendofl986, ecalclecionbyalandslide,andtheSuzukiCabinec was reamignedto the Japan National Raways gained a stable majority in the DieL In March SetdementCorporationalongwithanysurplusral 1981, the ProvisionalCommittee on Administm- estateofJNR and the shares in the nevwycreaed riveRtefrm (the ProvisionalCommitece) was ap- JRs. poinedundertheCabinesjuidictionandcarged 1.10 In addition to these administradve wih devising the drastic measuresnecessary to forms, the tax sysn was also reformedin 1988, achieve fiscal reDm wihout raisingtaxes. The and thc national consumption tax dtat had been CommitteewasheadedbyMr. ToshioDoko, hon- proposedearlie was implemented.These actions, orry chairman of the Federationof Economic coupledwithEfrvorableeconomicconditions,made Organizations(Keidanren), who had an eellent it possiblefor the governmnt to achieveits goalof recordofsuccessfizlyrebuildingprivatecntepriscs. discontinuing the issuance of special bonds by The Committeealso induded xpertsfrom various 1990,albeit sixyears later than originallyplanned. fields,including finance, government, and labor. The 1980s thus reprCsCnta transitionperiod be- 1.7 TheProvisionalCommittcediscussedhow tween the high-gowwheconomy of the postwar to achievean effective,flexible administrative sys- yearsand the present, stable-growtheconomy. As memthatwouldbeable toaccommodatechangesin such, itwasalsoaperiodofreevaluationandrefrmn social and economic conditions. In addition to ofthegovernmen's initive,financeandtax recommendingradical reforrn of Japan'sadminis- systms. trative system,the Committee also proposed the Ciis w JNR and its Causes privatiztionofJapan's three largestpublic corpo- rations-JNR, Japan MonopolyCorporaion (to- Developmentof Cris bacco and salt), and Nippon Telegraph and Telephone Public Corporation. They also sug- 1.11 Inl980,Japanhadapproximat*ly27,000 gsted tat privatc initiative be used to construct ldlometcrsofrail lineswhihwe-reoperatedjointy social infrastrucure, which in the past had bccn by JNR and more than a hundred privaterailway ahieved primarilythiugh public works. companies.With control of 80% ofthe totl num- 1.8 Basedon these recommendations,the fol- ber ofrail lincsthroughout Japan,JNR was by fir lowingsteps were taken;(a) all itemsthat required the largst nationvide operator. Pnvate railway Japanese National RailwaysPrivatizaton Study II 3 networks operated throughout Japan, with the rsulting from cssive capitl investment and largst onesconcentmted in the three mtropolitan interest acruing on tic large volume of debt. areas of Tokyo, Osaka and Nagoya.JNR was in JNR's consistentpolicywas to continue to coverits dimectcompetition for the commutermarket with deficitsby borrwing and, as a rcsult, itS mowldebt privateconcerns operatingin these areas. continuedto snowbalL 1.12 JNR had enjoyedvimtal monopolystatus 1.16 In spitceof its financiallyprecarious sieua- unril the 1950s. Gradually,however, competition tion,JNR continuedto maintainexcllent railway from automobiles, aircaft and coastal shipping engineeringand operatingtechniques, owing to its inecnsificdand JNR lost its competitiveedge in all historyofrecruiting officcrs and engineersof high geographicregionsexceptbothinthemctropolitancaliber and moralc.JNR's managcmnt remained areasand in the bullet train networks.But cven in confidentthatitwas theonlygroup capablebothof chemetropolitanareas,repeatedfarehikesinJNRs running JNR and of reorganizingit. JNR was finalyears as a public corporationrcsulted in fares opposeduntilthclaswmomenttotheGovemmcens that were higher than those chargedby the private proposedbreak up and privatiaon of the corpo- rail lines. JNR's sharein passengertransport vol- aion.The largestlabor unionwithin JNR which umedroppedsharplyfrom55%in 1955to 23% in wassocialist-led, also considered its role as part of 1985.Theshareoffigihrtansporcvolumedropped a politicalstruige and was actvely opposed to the more drasicaly from 52% to 5% over this same privatizationmovement.Butfiequentstrikes along period. Meanwhile,JNR's investuents in infr- with scuhing attck from the press turne the struucre only increased its debt burden. Road public againstJNR by 1980, lcavingthe general transport,onthe otherhand,beneficedfromfinan- impression that restucturing was essential Be- cal rsources obtained primaily fromthe gasoline causeof thesecircumstances, the implamcntation tax. of the prnvation processinvolved dt dismissal 1.13 JNR could not compete with the ocher ofmostoftheJNRmnagementandoftheincran- modes of transportauonwhich were being ino- sigent union's leaders (this union subsequendy duced during this period. Afiter 1964, the year collapsed).These were ather drstic masuresfor when JNR posted its first sine-year lossof $273 Japan. The other inority labor unions wihin million, reconstruction plans were formulated, JNR chose to cooperatc with the prvtization abandonedand then replacedby new plans These process. attempts at reform byJNR managementall cnded in failuredue to mistaken assumptions,lack of Reasonsfor JNRs Manageria FailurTebefore senitivity to the changingtranportation environ- Rewauring)foPr1 uziauo ment, and delaysin approvalby the Ditr. 1.17 JNR's manageral failuresprior to privati- 1.14 In the 1970s,JNR5s manag"'unt ~'mto- zation can bc attnbuted to a number of causes, ducedalrge-scaleproducivityimprovementdtive classifiedinto the three categoriesbelow. for the workforce, but the largest labor union within JNR dlaimedthe driveto be an unfairlabor a. Problemsinherent in the public corporation practedesignedtoundemine theunion.Someof systemsuch as. the daimswereupheldbyagovernmentarbitmtion i. Contol of the Goverment and ineven- agecy,andthemassmediasupporcedtheunionby tion of outsiders.In reurn forJNRs mo- cinazing the dnvec The productviy iLmprove- nopolystatus as a public corporation ment drive cnded in total defeatfor magement, providinga keypublic service,it was sub- resuling ina negligiblereduction in the work force ject to stict governmentl and congres- and an unprecedenteddeterioration of employee sionalsupervision and controL For discipline. ezample,the approvalof the Ministr of 1.15 Perfonnancecontinuedtodeclinethough- Transportwas required in order to con- out the 1970s and early 1980s. In 1985, JNR's struct anynewline and theJNRbudget annual loss before subsidiesapprached $22 bil- had to be appwrvd by the Diet .As a re- lion. The causesof the deficitwere:(a) abnormally sudt,outsiders, induding politiciansand the high personnel expenses(amounting to 78% of Government,ofien interferedin its man- cotalrvenues at their peak, comparedwithabout agementand forcedJNR to investin un- 40% for private railways),and (b) the burden pofitable routes. 4 Japanese National Ralays Privatizanon Study II iL Lack of incentivesfor efficientmanage- failed to improvethe profitabilityof the cor- ment. JNR managementmade littl cffirt poration as a whole. to maximiz prfits by rducng costs and ramisingproductivity. This behaviorcan be attributed to 1) a lack of managerialdiscre- . m . . . uon, itsef the resllt of considerableoutide interfrrence,and 2) the absenceof profit- 1.18 InJuly 1985, the SupervisoryCommittee motivationunder the publiccorporation submitted its written opinion concerning the r- system. scturing ofJNR to the Prime Minister. InApril ..i. Slow re.ction to intensifiedcompetition. 1987, as the first and most important part of the iiLSIowreaction tO intensifiedcompetiron...... Publiccorporations generally function n process,the corporatzion and divi- .ffic,velyin a monopolisticenviroanent. sion ofJNR werecaLrried out vinually in line with ,, , ,...... the Su D asxCommttee's re rt. The main However,as a resul ofa rapid incrcascim th..Su.iso.Commitees.port. - - competitivemodes of transportationsuch resultsof these stepsw : as automobilesand aircrfis, the Japanese a. The passengerrailway division was divided railwayindustry gradually lost its monopo- into six companies,threc on the main island listicposition and market share.JNR man- of Honshu and one each on the three islands agementcould not cope with this situation of Hookkaido,Shikoku, and Kyushu.A sev- without having the power to take appropri- enth firm handledfreight for the cntire coun- at m easures,sch.as the swiftelimination tay. The profitabilityof each new company of unprofitablelines. was carefullynasured andsteps werc taken iv. LIaborunions' lack of smns of crisis.Even so ensuaenagt thbsecompanics teineiga afterJNR began to run at a deficit.the solid mual bases.Ile thre biggr laor uniots refusod apprciaa the scale Hoshu-based companiesand the figi: of dne cdsisand continuedto puwe politi- company...... assumed only S42bilion of lNRs caissues. Thbeyalsor sedt cooperate...... ootal debt. The Shunanse (Bullettrain) with.manateye's effrts twrauonalize Holding Corporaton was establishedand aih rmauce mset's atfitud rs e in a sumed$dS52 billion ofJNR debt. It leased ae d c ismptionof wok-placc displine and bullet train netwodrsto Honshu-basedcom- a diorarion io th e ofiuaIityscrip an panics with the rcntal &fcsadjusted to correct the profit disparitybetwccn old and new bul- b. SizeofJNR and insensitivereacton t local let train networks The smallerpassenger com- needs.Ihe nationwideconcern was simply panies in Hokkaido,Shikoku and Kyushudid too largc to be adequatelymanaged by a single not assumeany of the JNR debt. The Man- managrial entity. Those in the railwayindus- agrnent StabilizingFund was created for ry must be sensitiveto localdemands and these smallerpassenger companies to oflfet provideconvenient services for dthircustom- the imbalancebetween rcvenuesand expendi- ers. Under unified, nationwidemanagement, tures. The Japan National RailwaysSecde- JNR was unable to provide locally-basedser- ment Corporation (JNRSC)provided the vice and consequendylost its competitive capital for this Fund. positionvis-a-vis the private railways b. The remainingS232 bllion liabiities were c. Lackof profit from related businesses.Given asumed by the JNRSC, establishedas the its pubLicnaure, the raiNwayindusty is regu- successorto the old JNR It also received lacedin dtermining itSfares. Even private sharesof the JRs and the surplus real estateof railwaysdo not genraUllyexpec high profit JNR as well as the redundant workes Thc levels.Indeed, related businessessuch as real redemptionof debts assumedby the JNRSC estatedevelopment account for a substantial was tO be achievedas far as possibleby the proportion of the total profit of most compa- r ent of dte ShinkansenHolding Corpo- nies.As a publiccorporation with specific ration, salesof the marketablepieces ofJNR- objectives,JNR was limited with respect to owned rcal estate,and salesof JR stocks.The the scopeof its business.Thus it could not remainingdebt is expectedto be born by the rly on rlated businessesfor profits,and generalbudget account. JapaneseNational RaihwaysPrivatization Stdy II 5 c. JNR had about 200,000 redtundantworkers in list these companics and to sr seling share 1980.By 1985,the numberoflNR employees beginning in FY 1992, completingthe saleof ali had alreadyboen reducd to 280,000 (through sodcs within four to fiveyears. However, the swok prior rstuauring) comparedwith 414,000 in narkerwasdeeplydepressedthroughout1992 and 1980. However,there wercstill an estimated concemswere raised that the offeringofJRs'stocks 93,000 redundantworkers. Thc Govcrnment to the public in a massivcscalc would further acively promotedvarious measurcs to securc damagethe conditionofthenmatker. For thisreason job changesfor these workcrs.The public the Govenmmt postponed the iitial sales of scaor was asked to employJNR workers, sharesuntil September1993 inspiteofthe desireof without increasingtotal employees,and the these dhrcecDmpanies for an earlir offeing. privatesector volunuuily accepted someJNR 1.21 In September 1993, 62.5% of JR East's wo . The eliminaon of the redundant shars were ofl red to the publicrThe remaining work forcelargely by job-tmnsitionwas car- sharesofJR East,as wll asshares ofJR Centraand riod out smoothly,aided in part by economic JRWest, arc scheduledtobe soldfrom Fiscal 1995 growth and by the very low unemployment onwards.The remaining three island companies rare of the Latc1980s. As a result, the number have pooreroperational bams and havebeen prof- ofJNNRemployees ultimately transfrred to itable only duc to the added revenue from the the JNRSC in April, 1987 was 23,660, imost ManaemenStabilizingFund,buttheyieldofthis of whom found jobs in other placesby April, Fund has been lowcr during the past two years 1990.A total of only 1,050 cmployceswerc becauseof low intcrest rates.Among thcm, how- ulimately dismssedbecause dty refusedto ever,JR ShiCOkuhas already met the required moveto newlyallottd placments. profit smndad. JR KyushuandJRHokkaido have d. Chaimen of the new JRs wer chosenfrom nortyetmet thisstandard. The current profit ofthc among influentialleaders in the localbusincss freight company is still belowthe requied kveL community.The first PreskientsoftheJR These thre ilandcornpaniesandthefreightcom- Eas and JR WcstCompany, the two largst panyarc alsostriving to achievethe ealiest possiblc of the JR companies,were former high-rank- listng, but the time schedulkis uncertainas ye. ing officals of the Ministy of Transport and P foiowig JNR had been supportrs ofpmrnazarion.The Presidet of the otherJRs were selectedfrom 1.22 The performanceof JRs sincerestructur- JNR managementmembers who had not inghadbenremarkablekespeciallyinthepcriodup opposedprivatzation. Otherlcadcrs within tO 1990, thanks party to the economic growth JNR who had promoted privatization(though which occurred in Japan during the period when as a minoriy at the beginning)were appointcd restruturing wascaried out. The annual inse to influential managementposts at the newJRs. in passengersand cugo was5% and 10%, respc- tidvely between 1987 and 1990. Thc number of Salesof Shares employeeswas drasticallyrduced to 191,000 by 1.19 SinceApril1987,theJRshaveoperaredas 1990. As a result, transportation efficincyrhas comerdalized stock companies, but all of tie improvedremarkably.Theprofitabilityofthenew sharesexcept 62.5% ofJR East'sshares whicl were JRs has been improveddue to the reductionof the sold to the public in September, 1993, wcrestill wok-force and liabilities.Transporation volume owned by the JNR Settlement Corporation. As has been increasedby JRs' own efforts and by soon as the JRs met the listingrequirements of the favorableeconomic environmntL The amount of Tokyo StockExangc, teir shareswere expected annual profit improvementfor the new JRs as a robelisredandsoldtothepublic,andtherransition whole is cstimatedto have reached$25 billion in to privateentcrprise would be completed. 1989 ( exduding the contributionof the favorable 1.20 On April lst, 1992, all of the JRs cel- economic environment). Of this improvement, ebrateddtir fifih anniver, one of the rcquire- $12.5 billion in profitsarc estmated o be due to meats for the lisdng.By chisdatc, three of the JRs efftciencygains resultingfrom restructuring.2 La- (JREast,JR Cental, and JR West) had alsomet all 2 SeeKoicikoFukui.JapaeNar"?dAiIwayP?wiiff- ofthe other requirenents.The Govenment's and tm&v4 7. ew*fJ*=xdL"wj6r&diV- theJNRSealanent Corporation's objectivewasto ig G ieWorldBank Dicussion Paper 172. 6 JapaneseNational Raiways Privatizaton Study II bormanar -ewm lonshavealsobeenimproved. Chaper III). The performanceand oudook ofeach JRs areacively expanding heirbusinesses to areas JR affecesthe rime scheduleo fstocksales needed to outside tasportton, as a resultof dreguladon, finalizeprivatization. and their oDrporateiage has impwved dramat- Rem2nin Issues caly.ReIs s 1.23 While the JRs have demonstmted better 1.26 With the creationof theJRs in 1987, the performance,thebebeneficiaryofdis restruc- firstphasoftheJNR privatizationwascompleted. mring is the Governmentwhich benefitsfrom the Ar that time, the followinglongero-tern issucs re- reductionofits fiscalburden. Beforcreructring, mained to be resolved.(See Chapter II.). annual subsiies from the Government to JNR Debt o reached about $5.5 billion on averge. After re- flASC structuring, the annual subsidy to jRs as been 1.27 A major portion ($232 billion) of JNR's drasticailyreduced. In 1991, for emmple,the sub- liabilitieswerereassignedtothe JNRSC;longwith sidiestoJRswasonlyS 1.0billion, indudingthat to therealestaeandsharesoftheJRs.Teseliabilities the JNRSC. This year. profitableJRs paid $4.0 are expeatd to be repaid by the sale of real estate billion in tax and contributionsto the Govemn- and shares.just after the fonation of the JNRSC ment, rcsulting in an overal surplus paid to the in 1987,thepriceoflandinJapanbegan tosoarand Govcrnmentof$ 3.0 bilLion(See Chapter M). greadyappreated during 1987 and 1988. How- 1.24 Rail passnges have also benefitedfrom ever,the land saleshave been carriedout at a much an improvement in servicequality (without fare slower pace than initially planned owing to the increascs).On the other hand, cerain segmentsof Government'sdecision in October, 1987 to pro- the soacEysuch as cquipmentsuppliers and other hibit competitivebidding for the rcal estateof the prvatc railwaycompanies face more competiton JNRSC The Governmentmade its decisionat the from JRs' commerciation. Passenges in the cxpenseofandagainstheobjecvsofdheJNRSC, separted remote ines pay highr faEs in some becauseit wasuthough tat such biddg might cases. However.there have been no obviouslose push genal land pricesup firthe. Therfrc, thc xceptforthefolrerJNRmanagemenrandradical sale of real este has been limited.The sale ofJR union leaiers,as wedlasthe 1,050employees who shares has becn deayed longr than inially ex- wre dismnissdin April 199. The privatizaion pectd due to the depressedstock market. processofJNR can bejudged to havebeen success- 1.28 As a rcsult, the prceeds from the salc of ful, even though the JNRSC still has outstanding assetshave not covered interest payments,which debts otaled$56 billion betweenFY1987 and 1992. In 1.25 Since 1991, with the prolonged and in- fact, theliabiLitieshad inarasedto $242billionby creasinglyscriouseconomic recession inJapan, the the end ofthe FYI992. In 1991 land pricespeaked parformance and oudook of each JR has begun to out and showeda sharp dclinc subsequenndy.Ac- differ. In 1991 JR passengerfigures still showeda cordingto figurcsrelased byJNRSC. the valueof demandincreaseof4.9%,butJRFrcighe's emand its land (basedon the posted price ofApril 1993) for cargo eveledoff In 1992 the demand increase was roughly$76 billion. Therefore,the Govern- ofJR passenge as awhole slowedto 1.0% andJR ment hasliftedthe ban on competitivebidding,but Fright'sdmandshowedadedincofl.9%.Among the oudook for land pricesis stillbleakand salesof JR pasnge this year,JR Central (whichreles Corporaon's land are sll slow inspite of various heavily on business passengers of Tokaido measuresto promote land sale (Seethe next sec- Shinbasen) andJRShikolk (whichfaces tougber tion.).Therepaymentofdebt,tiemajor remaini competitionfrom newlyopened highways) epen- issue to be solved. must be caried out rapidly enced dedine in pasengers for the first tme. JR through the sale of shares and land. or by the Freight facesincreasing compeition from tuckm, maxiiaionofprofitthroughthedevrlopmentof and is alsosuffring from recession.The recession real estateholdings. continued in 1993 and thre passengerJRs (JR Ucmaixy cernwg Sala qfjRs'Sha Hokkaido,JRCcntmlandJRKyushu) exienced a slightdecline in pasenger gwwh. This year the 1.29 Assoonas theJRs meet delising require- growthoftalpassengerjRsasonly 0.1%,while ments of the Tokyo Stock Eang:I their shares JRFrmightregisceedanotherdedineof4.5%.(See were exeted to be listed. Three of the JRs (JR JapaneseNational RailwaysPrivatization Study 11 7 East,JRCcntralandJR West)hadalrcadymet the public, but a substantialportion of chse shares requirementsby FY1991 and the sales of their remained unsoldbecause the sellingprice was too shareswas planned to begin in FYI992. Th-ebasic high. Subsequendythe Governmentdecided the principleof the JNR SupervisoryCommittee that salesofJR West's sharesshould be postponeduntil all the sharesofJRs be listedand soldto the public Fiscal 1995.The Governmentconsidered that the has not dcanged, but the time schedulehas been absorbativeability of the stock market for priva- delayed for fear that the massiverelease of new tizedpublicompanicswasnotbigenough. Owing stocksmight furitherdepress tealready weakstock to this decision,the completionofthc saleof shares markeL in the thrde Honshu JRs has been delayedstill 1.30 InSeptemberl993,62.5%oftheJREast's further. Shareswere offcred to the publicfor the first time. 1.31 Regardingthe seling off of the stocks of Theremaining1.5 millionunsoldJREastshares,as th threc IslandJRs, in addition to the weakening weilasthose oftheJR Cental (2.24 millionshares) of the companies'profit base(due to fallingyields andtheJRWcst(4millionshares),werescheduled of the Managment StabilizingFund) the argu- to be soldfrom fiscal1994 and later.In December menstillsrmainsunsolvedastowhetherthelisting 1993, the Vice Ministcr for Transportation an- ofacompanythatisnorderivingoperatingprofits nouncedthat only one ofJR West and JR Central from its main businessis appropriate,or not. The wouldbesoldoffinfiscalyear1994, indicingthe timing ofthe disposalof the three IslandJRs'stock abandonmentof the intention of thc Ministy to is thus uncear at presen. listbothcompanieson the stockexc;magesimulta- 1.32 ConcemingJR Freight,in addition to the neously.InJunc 1994 theAsset DisposalCouncil need for maintining a certain Ievd of profits in ofJNRSC conduded tht onlyJR Wcstshould be order tO achievc lisdng on the exchange,most listd in Fiscal 1994. This decisionwas made in analys agreeth many questions remain un- light of the postponementof the sellingoff and solved,such as cost allocationwith passengerJRs, listing of Japan Tobacco Inc (another Govern- the burden of new investmens,competition with ment-ownedcompany) from fiscal 1993 to fiscal trucks, resrictions on cpnding business, and 1994. In October 1994, one third of Japan othe This rendersstodc disposl difficult in the Tobacco'sshares were offered tO the bidders and near futurc. IL RecentDevelopments Affecting the RemainingIssues

Issues of the Japanese National Railways sharcsofthc Tcito RapidTransitAuthority, which Settlenent Corporation operatesdhe majoricyparts ofthe subwayin Tokyo. 2.1 OneuniquechacerisdcoftheJNRpriva- These variousfinancial resources were allocated to tr:auion is tie way JNR's debt was allocatedto JNRSC to financedebt redemption. various bodies. The debt induded the liabilities 2.3 The JNR-rclated long-termdebt requir- accruingin the future, such as the expectd fumre ingsealementtotaled $337 billion(V37.1 trillion), deficitsof the three sand companiesand fu indudingthe liabilitiesacauing from such bodies pensionpayments, as wel as the enormousamount as JRCPC, the Honshtt-ShikokuBridge Authority ofoutnding debt. All this reflectsthc principles and fiture pensions,in additionto the direa $227 of the Supevisory Committee for JNR Recon- billion of JNR's debt. Of dtis $337 bilion of stuction, thatthe railmadsector must continue to liabiLities,JNRSC inheriteda total of$ 232 billion, play an important role in the fiture and newJRs induding fture expensesof $52 billion such as must havea soLdmanagnia basis,that the burden pension liabilities.But this figuredid not indude to the nation must be minimizeO!,andthat thisype future interest payments on its liabilities,whih of restructring should not occur again. As an werc cakulad as $72 billion. Four JRs (East, important measureeo achievethese principles,the Central,West and Freight),the ShinkansenHold- JapaneseNational RailwaysSetdement Corpom- ing Corporation (SHC) and JRCPC inherited the tion (NRSC) was crated and assumedthemajor- remainingof $105 billion of liabilities. ityoftheJNR'sdebtaswelUasthesurplusassecsand 2.4 Of the $232 billion inheritedby3NRSC, the sharesoftdenewJRs. Sinceonemajorproblem. atotalof$26bilion wasscheduledtobepaidinrthe ofJNRwas the existenceofthe accumulateddebt, time period up to fiscal ycar 2016 by the thre thecvaluation ofJNRSC constitutesthe important HonshuJRs (JR East,JR CentraLandJRWest) for fiactor of the JNR privatizaton itsclf. the rent of thc Shinkansenficilities through SHC. TDefNCorc,the amount of dcbt the JNRSC itself Debts .fNR,SC had to redeemwas $206 billion. 2.2 JNRSCwascratedprimarilytorepaythe 2.5 ITeFourJRsinherited$42bilionoftheir long-termdebt held by JNR through the salesof own debt and, in addition, had to pay $89 bilion, surplus asset and of newlycreatedJRs' shares. To orginally inheritedby SHC, JNRSCandcJRCPC, this end, folowing theApril 1987 prnratizationof making the total $131 billion. JNNR,it recived 8,180 hecres ofland fromJNR, 2.6 Since their birth, these four JRs have re- equal to 13%ofth 65,380 hccrcs ofJNR's total paid some debt, but ar the same time undrtaken land, and 630 hectaresfrom the Japan Railway new borrow:ng.So, itis not dear how much ofthe Consauction PublicCorporation (JRCPC),total- originallyacquired $131 bilion liabilitiesstill re- ing 8,810 hectarcsof land, as weli as the entire main. However,according to a report by the Min- shares of the new JR companiesand about 54% isiyof TransportationinOctober1992, theoriginal 10 Japanese Nationl RailwaysPrivatization Study I1 mcntioned above,$26 billion,which will be paid sales.Althoughdebtredemptionby the endoffiscal by three Honshu JRs, was to be passed on to the 1992 amounted to $61 billion, new debt was ShinkansenHolding Corporation (reorganizedin added. totaling $56 billioninduding interestpay- October 1991into theRailwayDevclopmentFund) mcnts, plus,pension and othierexpeses, leavinga in the forrn of a loan fiom the JNRSC to that debt balanceof $242 billion.Out of this amount, company,$70bilionwastobeeamedfromthesale $17 biDionwill be collectedfrom three debt JRs of land, and $11 billion from the sale of shares through the RailwvayDevelopment Fund. So, (calculatedat a fice valueof$5 billionfor thcJRs' JNRSCactuallyowed$225 billionat the cndofFY stocksand at revalued$6 billionfortheTeito Rapid 1992.Thisamounis$19 billionmorethanJNRSC Transit Authority)for an cnvisagedtotal redemp- inheritedat its birth. However,this amount would tion of $107 billion.The remaining$125 billion be $167 billion without intcrestpaymcnt- $39 was to be financedby the GeneralBudgetAccount. billion lessthan it inherited. This was supposed to be the final burden to the 2.10 By comparing the amount of subsidics nation. However,the calculationdid not indude receivedfrom the national govemment with thc fuEureinterest on JNRSCs liabilities, amount of taxes and contributionspaid by JNR 2.9 Following this setlement in 1987, the and succeding insitutions to the nationalgovern- JNRSC has proceeded with debt redenption mentandtomunicipalbodics,onecmyobservcthc through the salesofland, 54% ofthe equity in the burden borne by tax-payersbefore and after the Teito Rapid Transit Authority, and other asset privaization ofJNR.

Table l-1: Trendsin he Long-TermLiabilft of JRs (USSbillions) Beg4ming; EndofFY of 1987 1987 1988 1989 1990 1991 1992 JNRSC 206 211 219 228 221 223 225 JRs 131 128 122 117 '115 115 na JRstemselves 42 40 35 34 32 31 n.a. owed to HC& JNRSC 78 78 76 74 74 74 na. owedcoJRCPC 11 10 10 10 10 10 na. Total 337 339 341 345 336 337 n.a SShinkansn Holding Corpora ion was reorgaized as the RailwayDevdopmenc fund in October1991. Note:In this able.final debors (payers)are shown. The daa derore diffrs fiomta for inhried debtos

Table11-2: Taxes, Contibutions andSbsuidies (USSmjlions) FiscalYear Taxes and Contn$utions Subsidies Balance 1982 320 6,631 -6,311 1983 348 6,380 -6,032 1984 421 5,885 -5,465 1985 435 5,455 -5,021 1986 455 3,433 -2,977 1987 2,006 1,761 245 1988 2,234 1,918 315 1989 2,092 5,755 -3,664 1990 2,722 1,428 1,294 1991 4,039 984 3,055 NotcForcxecontributionsandsubsidiesfolo6wingtheprivadntionofJNRfigspretanozlofdwsefromtheindiniduaI JRs.JNRSC. the ShinknsenHolding Corpomtion, and the RailwysMainenance Fund. Japanee National Railwys PrivatizationStudyl1 11

2.11 Thc Ministsyof Transportation rvcealed 1986, total lossesamounted to about $73 billion, such figuresin October, 1992.According to thcse, but in the fivc years followingprivatization fiom bceforepriv.-tization, the amount of subsidiespaid 1987to 1991,despite the fact that the sal ofland to JNR far exceededthe amount of taes and and stockshas noc proceededas smoothlyas was contributionsreceived from JNR (as a public cor- foreseen, totll losses have filUcnsharply to $18 poration, JNR was not required to pay frderal billion, includingthose of the JNRSC. taxes).It wasestimated that the administationof Asst .f'AFSC JNR imposeda burden on tax-payersin the order of $5.5 billionannually. 2.14 Outoftheassets that theJNRSCinherited 2.12 Followingthe privatizationofJNR, how- as financialresources for debt repayment,by the evcr,the individualJR companieswererequired to end of fiscal1992, it had sold fixedassets, such as payfederaltaxes such as the corporatetx and the 3,917 hecares of land, plus rail cars and other amount of such taxespaid was quite considerable, assets, for a total of $29 billion. As a result, the reflectingthe favorable. nagementconditions of amount of land held by the JNRSC fell to 5,314 these companies.The amount of such taxes col- hcars On the other hand, 421 hecarcs of land lectedcame to exceedthe amountof subsidis paid havebeen returned by the JRs becausethey do not to JNRSC etc, implyingthat the privatizationof need them for their activitics.At the time the JNRhasbmughtconsiderablebenefits thejapa- original debt redemption plan was formulatd, ncsc tax-payingpublic in subsequentyears. (In jNRSC had planned to repay $70 billionthrough 1989the amount of subsidiespaid by the govern- the eanings from land sales(thc valueof the land ment exceededthe amount of taxesand contribu- that couldbe sold off from the towl8,810 hectares tions received, due to special pension-related inherited). reasons6which will not occur again.)In 1991, the 2.15 Immcdiattly afier the formation of the net balancefiom JNR heritance to the govem- jNRSCin 1987,thepriceoflandinJapan beganto mintrached $3.0 billion. soarandgreadyappreciaedduring1987 and 1988. 2.13 Moreover,ifwelookattheoverlbalance By the beginningof 1990, the appraisedvalue of sheerofJNRveedbytheMinisryofTranspor- the land hed by the JNRSC had rise to about rtion, in the five yw period fiom fiscal 1982 to $136 biDlion.During these yas land saleshave

Tablel- Finncial Sta of JNR(FY 19812 - 1986) and JRS (1987-91) including JNRSC (LIS nillions) 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 Opraing Revenu- 26,812 26,915 28,206 30,095 31,062 32,117 34,554 35,907 38,588 40,140 % cbaagefiwm prnviamperiod (4.2) (0.4) (4.8) (6.7) (3.2) (3.4) (7.6) (3.9) (7.5) (4.0) Opertniu Expense 35,249 36,701 36,330 38,420 34,298 29,784 30,710 32,195 33,077 35,418 PmaxdEi,w (18,704)(19,219)(19,035) (20,931) (19.229)(16,220) (15,563) (16,811)(15,862) (16,470) Openg Profit -8,436 -9,786 4,125 8,324 -3,237 2,334 3,844 3,712 5,511 4,722

Profit 7,243 7,909 7,349 6,980 6,467 3,411 4,635 9,083 17,645 9,640 GuuNSfiomSalt oJ'fivdA=& (643) (1,51C) (1,392) (1,404) 2.927) (1,207) (1,850) (2,265) (7,373) (7,065) EordinyLoss 307 371 287 279 926 590 1,038 1,238 1,940 2,107

Sourc:Minims ofTanspot 12 Japanese National RailwaysPrivatization Study I beencarriedoutat a much slowerpacethan initially billion,outofwhich $5 billionwouldrepresent the planned, owing to the Government'sdecision in shamesof the JRs, calculatedat facevalue. Had the October, 1987 to prohibitcompetitive bidding for threeHonshuJlRsbeerlistedonthestockexdange Ehereal cstatcof theJNR SerdernenrCorporation. at the peakofthc so calledBubblc Economy, prices The Governmentmadc its decisionat the cxnse as high as Y1 million per sharc, twenty times as ofand againstobjections from theJNR Setdement much as the face value, were suggested.At this Corporation, because it was thought that such price, the totd 8.24 million haresof dhesethree biiddingr.aieht push generalland pricesup fiurther. companieswould have fctched$75 billion. How- Therefore,JNRSC missed thc opportunityofsell- ever, given the collapseof the Bubblc Economy, ing real cstatesin an exemely fiavorableenviron- wereallte sharcsto be soldin die prcsentsiuation ment. of weak share prices, (taking the Y380thousand 2.16 In 1991landpricespeakedoutandshowed seling price ofJR East sharesas of August 1993), sharp declinesubsequendy. According to the fig- the figure for the three JRs would. be only $29 urcsreeasedbyJNRSC,tevalucofthislandbased billion. (Actualsale of JR East shaec: 2.5 milion on the postedprice ofApril 1993,was roughly $76 forY1 trillion;JR East remainingsharcs: 1.5 mil- billion.Thcoutlookforlandpricesissdllblcakand lion, JR Central: 2.24 million shares,JR Wcsr. 2 the salesof JNRSCs land is still slow in spite of million sharcs, total 5.74 million sharcs times variousmeasures topromoteland sales(See follow- Y380thousand = $20 billionor V22 trillion) ing section.). 2.18 Thus, ifjNRSC werc to sellall of its land 2.17 Similarly,regarding the sales of stocks,the holdingsat the posted price ofApril, 1993, as well originalplan assumedthe salesrevenue to be S11 as all th sharesfrom thc dhrceHonshuJRs ar the

Table114: Trends in LandPrice (Cmrdnal Distict Peent Chag fmmPeceding Yer (Percent) Area 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 Tokyo 5.5 7.2 12.5 48.2 61.1 3.0 4.8 4.1 -6.9 -19.0 (106) (113) (127) (189) (304) (313) (328) (341) (318) (257) Osaka 3.9 5.0 7.0 13.2 37.2 35.6 46.3 8.1 -19.5 -24.2 (104) (109j (117) (132) (181) (246) (360) (389) (313) (237) Nagoya 2.7 2.7 3.3 6.4 16.8 21.0 22.4 19.1 -7.6 -13.7 (103) (106) (109) (116) (135) (164) (201) (239) (221) (191) Local 2.8 2.6 2.5 2.9 5.4 7.6 15.4 16.3 0.4 -5.6 (103) (106) (108) (111) (117) (126) (146) (169) (170) (161) Note Thenumber in pamndxhsicde pricein 1983- 100

TableMl. Trendsin SharePres (yen) Ng" N;PPon Tdegapl Tokyo Electric Seibu StockrAvg. and Telephone Co. Power CO. Ra&IWay FaceValue (50.000) (50) (50) April 1,1988 26,104 2,370,000 6,050 4,300 April3, 1989 33,042 1,520,000 5,950 5I=0 April 2,1990 28,002 1,080,000 3,600 4,850 April 1, 1991 26,007 1,020,000 3,810 3,800 Apil 1, 1992 18,581 615,000 2,800 2,200 April 1,1993 19,099 977,000 3,670 3,000 JapaneseNational Raihwys PrivatizationStudy II 13 price given abovefor JR East shares (V380thou- wasactivelycontinuingitseffortsr oseUthe remain- sand), it would be able to pay back some $105 dcr. billion (VI1.6 trillion)-only about half of the 2.22 Accordingto the regulationsgiven in the long-term liabilitiesoutstanding as of the begin- JNRSClaw, in orderto ensurea fairand appropri- ning of fiscal 1993.This would stillleave ro-ghy ate exeution of the land sales,such salesare to be $118 billionof unpaid long-termdebt, the repay- carriedout by competitivebidding. In the caseof menc of which would have to be borne by the supplyingland for public use,official use or public General BudgetAccount. utilities,the sale may be conductedaccording to 2.19 At the peak of the economic boom in frec contrax, as stipulatedin JNRSC Law. 1990, had all the land been soldoffwithout being 2.23 However,at the time that theJNRSCwas constrained by regulations at the current price formed,land priceswerc alreadyappreciating rap- given above, and had all the shares of the three idly,and, in rcsponse,the EmergencyLand Coun- Honshu JRs been sold at VI million per share, termeasuresPlan was enactedby the Government which was then suggested,a total of more than in October, 1987. Undcr this plan, in those areas $200 billionwould have been made availablefor where the price of land was alreadyapprcaating debt redempton. Thiswould havemet the liabili- beyond reasonablelevels (apart fiom land being ties JNRSC had to pay, so that a much smaller supplied for public or officialuse) saleswould bc amount than expected would be borne by the defrred, and, in addition, the systemwhere land Gcneal BudgetAccount. couldb- disposedof without the disdosureof land Land Salesof.jRsC priceswould be broughtunder review.Accord- ingly, a study for land disposalmeasures without 2.20 The rcsultsof salesof land, train cac disdosng land pLicewas cared out and recom- and other assetssince the inepuon ofJNRSC are mcnded severalmeasurcs shown below. givn below.For a number of reasons,the sale of Pol taseRegdxsom C4xcn=gdxDipesd land snce the inception of JNRSC has not pro- gressedsmoothly.F=stJNRSCneeds to takelmany fL steps before it can complete salesof land, such as 2.24 Accordingto the JNRSC Law, land dis- coordinationwithlocal govement and ncibor- posalduough fe contractwUS to be allowedonly hood peoplc,paradarly in cascof large scal land in the following iune. in urban ars. Second,as part ofthe governments i In the caseof the transferof land t the state policyagainst the spiralingprice ofland, the usual or corporationsfor public use, officialuse, or system of Open Competitivc Biddig was e&ec- public utlties. tvely frozen. Third, dcmand for land has fallen ii. In the caseof the transferof land to localgov- sharply because of the recent recessionand the ernmentsfor public use, officialuse, or public dedining trend in land prices uilities. 2.21 As a result of land disposalsup to fiscal iii In the casc of the twansferofland to public 1992, the JNRSCwas still in possesion of 5,314 enterprisesfor public usc, officialuse, or pub- hectres oflandat the benning offiscal1993,and lic utilities.

Table116: Residts of Salesof FLxedAssets by te JNASC (USSWlli) land Method of Sale FY (hecares) Opeu CompetitiveBidding Free Contract Other Assets Total 1987 271 194 1,013 1 1,208 1988 966 255 1,595 5 1,855 1989 519 335 1,929 252 2,516 1990 810 212 7,161 4 7,375 1991 593 205 6,861 255 7,320 1992 758 301 7,824 73 8,197 Total 3,917 1,502 26,382 589 2S,473 14 japaneseNational RailwaysPrivatization Study II iv. In Ehecase of the transferof land for use by of immediateconstruaion on corporationsin arcasof businessthat had that and. been recognizedby thc CompulsoryPur- September1992 Easingof rcgulationsconcem- chaseof Land Law,or in the caseof arbitra- ing businessand commercial tion basedon the CivilConciliation Law. uses of land. v. In caseswhere there are no biddersduring 2.28 Moreover, thc Government decided in competitivebidding in either the first or March 1993 to removethe regulationdefcrring the second round of bidding and alsowhen a sale of land through open competitive bidding. contract is concludedat a higher price than However,owing to poor economicconditions the had been originallyenvisaged. saleof land throughresumed open bidding has not vL In the casethat the successfilbidder does worked well. In addition, local govcrnmentsof not concludea contract and also when the manyctiesarcunwilingtopcrmitandsalethrough contract price is higher than that offeredby open bidding in their areas because they are still the successfilbidder. wary of a 9are-up in land pnces. viL In the caseof the saleof land where the pwice envisageddo-s not exceed$27 thousand (V3 MeaSU for Land Disposa withoxt Dissming million). LJdPfic viii In the casethat competitivcbidding is un- 2.29 AtthetimeoftheBubbleEconomy,which s5i::bic. caused a rapid appreciationin land prices, there 225 In responseto the Emergencya ndcCoun- existedtheanxietythatsalesprices ofJNRSC'sland termcasuresPlan in October, 1987following r2pid would acceleratelandprice inlation in generaLFor inflationofland price, theAssetDisposal Commit- this reason,various methods were sought not to tee in May 1988decided to establishthe foilowing disclo the land price directy. An outline and methodswith the revisionofthe JNRSC lawregu- evaluationofthcse methods for land disposalwith- lating lnd disposalthrough free contract i) the out discosingland price is givenbelow. Sales of Developed Real Estate Method, ii) the S ofDevelopedRealEFsta Sma-ot Sals of TrustBenefiary RightsMMeod, 2.30 This involvesthe saleofland on whichthe iii) therLoan Convetibl into Real Esbta Method. JNRSC has erectedapartment bu:' ings or other Furthermore,based on decisionsmade by the am constructions,where the land and buildingare sold Committce in September 1991,in order to avoidt overlinfle. th pric of lad th copd,ogethecr as parEWaownership rights. As the sale overly inflatingthe p. ce of land, t compeive,e involvesnot just the land itselfbut alsoconstruc- biddingsystemwithacappricevasintroduced. h theland is sold is not ddition, in orderto facictart the sale oflarge-scale dear. thefich al was thadein the suburbsof real esratewithhigh assetvalue, theJNRSC lawwas revisedto include the Corporate Bonds Conver- TokyoinSep-mber 199l,andfirthersalsinvolv- ibleintoStocMethodPrepartionsto implment ing 54 8 properies had been made byJune, 1994. the newmechodsarecurrentlyinprogress.Further- ii TheSmailLotSalsefthe TrmstBenejiuayRtights more, a new schemecalled the Joint Salesof Land 2.31 This method involvesthe entrusting of and Housing with Private CompanyMethod was land to trustbanks inexcange for mst beneficiary approvedby the Committeein February,1994. rights,whicharesubsequendysubdWvidedandsold 2.26 As a resultofthe introductionof thesenew off to gencralinvestors through the mediation of methods, land sales through frie contract have tLustbanks The total sum of the sale of the trust increasedsince FY 1990,especially in Tokyo met- beneficiaryrights, whih is given by the present ropolimanarea. valueofthe trust dividendsreceived while the land 2.27 In addition to the establishmentof these is in trust and the money receivedfrom the salcof newmethods,thefollowingpolicishzavebeenalso the land and the buildingconsaucted by the rst adopted to caseregulations and facilitateland sales. bank after the tmst has matured, does not give a October 1991 A widening of the definition clearfigure for the price paid for the land itself of public utlies. 2.32 Foilowingthe first such salein December September 1992 Permissionfor localgovern 1989, a total of six sales has now taken place, ments to aqtuire real estate concentratedmainly in areas along the railway without the usual requirement tracks(In addition,one casewas cancled owingto Japanese National RailwaysPrivatization Study II 15 thewithdrawalofthe main tenant,andanotherwas companyare then issued.The redemptionof thcsc not completed due to the lack of subscribers). bonds is conducted by convertingthem into the However,in additionto the wakening of the real stockofthe saidcompanyafter dte elapse ofa given estatemacet, (owingto theamendmentoftaxlaws period of time. imposing a greaterburden on investorsthan be- 2.37 As the first sale of land under the above fore)the saleof the trust beneficiaryrights for each ystem,thesaleofland at Shiodome,locatedwithin of the piecesof real estatehas proveddifficult, and the Tokyo area, has been planned, and Shiodome land disposalthrough this methodis nor plannedto DevelopmentInc., thc company that is to receive extend it. the investmentin kind and developthe .eal sate, wascreated in October 1992.At present,the com- in Lo=wCox vemn intoReal EArt.te 2.33 In this method, f equivalent.ids to the pany is in the process of formulating concrete land developmentprojects costsarc paid in the developmentpla. formofa long-term,low-intcrest loan in retumrnfor IssuesConeerning Salesof JR's Shares the future right to joint ownershipof the pieceof 2.38 SinceApril 1987, newlycreated JRs have real estatein question.In return for supplyinglow operatedascommealized,commercialized-stock interest funds, after a certain period of time bas copanies,butailofthesharesexcept62.5%ofJR elapsed,the investoris able to exrcise the conver- Em's shareshave been owned by the JNR Setde- sion right, thus acquiringthe rightto joint owner- meat Corporation.The basicprinciple ofthe JNR ship of the real estate. SupcvisoryCommitteethat all the shars ofJRs be 2.34 Intheperiodfrom 1990toJune1994, five listedand sold to the public has not changed,but, such loanshave been receivedfor prime rcal estate as noted above,the timeschedulehas been delayed. in Tokyo and other areas from companies con- Inadditionsinctheco apVofthebubblecconomy sructing officesand commercialfacilities. In addi- and the advent ofa prolongedrecession, the opera- non, from December 1992 a system has ben tionaloudookforsomeJRshasdeteriorated.When stablishedwherebytheJapanDevelopment Bank, all the sharesare soldto the public, the transitionto a governcmt financialinstitution, supplies long- private enterprisewill be completed.Until then, term, fied-interest rateloans to subscribersto the the sales of shares continues to be an imporant joint ownershiprights. elmet ofJNR privatization. iv Joiat Sals of Land ad Honmingwitb Pice p.ies for the Sal ofJR'sShae

2.-35 Underth"Sales oflDvlopedRealEstat 2.39 The JRs, which are sixpassenger railways Methd,' private companiesare sdectd to con- an one t service,were bon fromthe pria- struct residentialbuildings on land hdd by the tiation and subdivisionofJNL Regardingthese JNRSC, and the individual residencesand land companies,theGovemmentdecidedto"transform belonging thereto are subsequentlysold of indi- them into privatelyowned companiesas early as vidua11y.Under this system, in the event that not all possible, followingthe fulfillmentofseeral condi- viduhissste, . Uner intheeven tht no aU tionssuch as the stabilizationof management."Its of the individuallots are sold,the privatedeveloper intntions, as explainedby the Minister ofTrans- will be requiredto purdhasethe remainder.Tlis enablesthcJNRSC to completethe disposalofthe porationand otherMmistrymembersto the Diet, land in question,even in the eventality of unsold wer as foll residences.As of March 1994 tWosites had bee a. Tkhegovenment would aim at the complete selectedfor this system,and the respectivepivate privation of the JRs though the sdelig off companieswere chosenby July,1994. of all shares,wvi are presendyowned by the JNR SettlcmentCorporation (thc JNRSC). a. CorpoateBond Gmorbl inxt Stock b. It would not be necesary for the Govenment 2.36 Thismethod,currenlyunderreview,con- to recaina controllingshare of the JRs, which cernslarge tracs of land with high assetvalues. A are plannd to bc priazed. companyis createdthrough investmentin kind in c. A basic principleof the privatization,in accor- thelandinquestion.Thesame companythenraises dance with Governmcntpolicies, is that the findsfortheconsructionofbuildingsandprojects. JRs be transformedinto pure privatecompa- Bonds with the right to future stocks in the said niesin order o act freely. 16 Japanese National RailwaysPrivatizadon Study II d. The Governmentis currendy studying dt 1989.For thisreason,theGovernmentandjNRSC timing and content of the amcndment or were forced co considcr limiting the volume of abolitionof the JR law,which will bc re- shares sold in considerationof the impact on the quired, if the JRs are to be privatizedthrough market. Regardingthe volume, the Government the saleof thc sharesand their listingon the decidedto sell off a half of jR East's stock As the Tokyo Stock Exchange. stock market continued to bc bearishthroughout 2.40 Followingthe reorganizationof the JNR August, 1992, the saleof the company'sshares was and subsequentestablishment of the JNRSC, the cventuallypostponed to fiscal1993 or later for the stocks issuedby the sevenJRs were transfcrredto fear of causing further disturbance to the stock the JNRSC. The number of stocks issued and market. capital of the JRs are givcnbelow. 2.44 In 1993, the Gavernment and JNRSC 2.41 As thestocksoftheJRswere designatedas agatndecided to carryout the sharesales operation important assets,serving as financialresources for for JR East,and the salewas implementedaccord- the repayment of JNR liabilities by the JNRSC ing to the followingschedulc: Law, the disposal of these stocks should require Agwt discussionand authorizationby the AssctDisposal * Salesof 600 thousand sharesby bidding. Council, a committer established within the * Pricing for open tender to generalinvesors JNRSC. In accordancewith this law, thc Council accordingto weAghtedavmge of tenderprices beganits deliberationin April, 1992 and prescnted ofabove-n tioned 600 thousandshares. its report inJuly of the sameyear, regarding the sell- Sepwber- off of the stocks of the three Honshu JRs, which * Applcation for the shares from generalin- alreadymet all the requirementsfor listingon the restors Tokyo StocKExchange. The stock of these three Oacobe JRswassubsequendyplannedto besoldoffaccord- * Allottingof Stocksapplicants seected from ing to the aboverccommendations of the report. above by lottery. 2.42 The requirementsfor the listingofeach of * Listingon the Tokyo Stock Exchnge dheJRsonth cTokyo Stock Echangeare shownin 2.45 W the sharcs werc offered to genetal TableII-8.AlloftheHonshuJRshavealreadymet investorsin Scptember, a total of 10.48 milL;on tesc requirementsfor the listing in fiscal 1993. bids were reccived for 1.4 mion shares to be Initidl SakgsofIR Eas6 S es released.Thercfore, buyers were chosen by loteq. When the JR East shareswere listed on the Tokyo 2.43 Based on the July 1992 report of the Socd Exchange,theywere populr beyondexpec- Council, che seling off of the three Honshu JRs' tation, necesitating the relcaseof an xtam570 stockswasexaminedconcretely.However, thestocc thousandshares that had been held in rve to act market at that time was in bad condition. The as abrake on the rising shareprice. The combined Nikkei Stodc Averageof major 225 stocks had revenueofthe inital bidding, theopen tender and come down to as low as 15,790yen in August1992, the sellingon the stockmarket ofthe extra sharesas less than half of the avcragelevcl of the peak year abrake rcached(fora totalof2.5 millionsharcs)VI

Table11L7: Summary of TheJR Stocks Par Value StockIssue Capita (thousand yen) (thousand) (billion V) JR Hokkaido 50 180 9 JR East 50 4,000 200 JR Cetral 50 2,240 112 JRWcst 50 2,000 100 JR Sbikoku 50 70 3.5 JR Kyushu 50 320 16 JR Freight 50 380 19 Total 9,190 459.5 Table11-8 Listing Requirements I Y Billion) _ _ Hokkaido East Central West Shikoku Kyushu Freight 3 Standards RequiredNet Assetsas of the end o of the precedingfiscal year 18 400 224 200 7 32 38

Stockholder's Net Assetsas of the beginningof FY 1987 259 297 166 155 104 311 34 Equity Net Assetsas of the end of FY 1992 264 544 399 260 120 320 46 (Net Assets) RequirementsQualification Yes Yes Yes Yes ; Yes Yes Yes 3 Profits Standards Three YearsEarlier 2.7 60 33.6 30.0 1.1 4.8 5.7 (the less profit of Two Years Earlier 2.7 60 33.6 30.0 1.1 4.8 5.7 currentprofits and PrecedingFiscal Year 3.6 80.0 44.8 40.0 1.4 6.4 7.6 profits after tax) FY 1990 1.6 107.8 117.2 60.5 7.4 3.0 6.1 FY1991 13 108.0 117.0 66.1 7.0 2.4 1.9 FY 1992 6.0 100.9 67.6 55.7 3.5 3.5 0.2 RequirementsQualification No Yes Yes Yes Yes No No Dividend Standards Dividingat the PrecedingFiscal No Yes Yes Yes No No No Year(Fiscal Year 1991) Other Standards OperatingMore Than 5 Years Yes Yes Yes Yes Yes Yes Ycs After Establishment Requirementson the listingfor TokyoStock Exchange No Yes Yes Yes j No No No Source:Minitry of Transport Notes:Nct assctsof theThrce Island JRs arc calculated aftcr deducting of theManagcemcnt Stabilizing Funds.

'-1 18 Japanese National RailwaysPrivatization Study 11

trillion. This amount far exceeded the V150.4 and JR Shilcoku)we expectedto make lossesin billion, the amount planned by the JNRSC as theirrailaybusinesss from theiniz p6n and,in revenucin its budget in the beginningof the fiscal order to covcrthe deficit,the ManagementStabi- year 1993. lizing Fund (Hokkaido Y682.2 billion, Shikoku 2.46 RegardingthesalcandthelistingofthcJR Y208.2billion, KyushuV387.7 billion) was allot- East stock,dissatisfaction was cxpressedover the ted to them. The return on the investmentof this fit that some major issuessuch as the fareregula- find is then used to cancelout the railwaybusiness tory system,which will affect the stock price, re- losscs.Thus, management'srcsponsibility in these mained unsolved.These are, in a sense,problems Island JRs' has been to restrict lossesmade from cxistingin any businessof a public naure, but in railway businessto within the amount generated thecaseofJRs the relationwiththeGovenment is from the investmentfi. d. Until 1989the Manage- the matter to be settled in the future.The issues ment StabilizingFuni was 1ent to the JNRSC, are likelyto affea the firturemanagement of the guaranteeing a yield of 7.3%, according to the company. scheme created at the time of JRs' creation.After 1989, each of the JRs would receivethe gradual TimecSchd/dcfi~rFuaureShartS^e &ilrepayment from the JNRSC to invest by itslf. 2.47 The remaining1.5 millionunsoldJREast Operting lossesof IslandJRs' in fiscalyear 1993 shares, as wel as those of the JR Centrd (2.24 decreasedfrom those of 1987 because of the in- millionsharcs) and thcJRWest (2 millin shares), creaseof opertng revenues,despite a decreasein were scheduled to be sold from fiscal 1994 and theprofitfromoftheFund. However,failingyields later. In December 1993, the Vice Minister for due to rece low interestrates, togetherwith the Transportationannounced that tre stodc of only limitation of incrcasesin the operating revenue, one ofjR West or JR Cental would be sold off in will weaken lland JRs' profit base and becomea fiscalyear1994,indicaingtheabandonmcnrofthe seriousproblem for theseJRs'. intentionofthe Ministy to list both companieson 2.50 Regardingthe sellingoff of the stocks of the stoCkCihange suiultUaneusly.This decsion the three Iland JRs, in addition to mesur con- was made in light of the postponement of the ceming the weakening of the companies' profit slling off and listing of Japan Tobacco Inc-, an- base due to falling yields, the argument remains othr Govenment-owned company from fiscl unresolvedas to whethr the listingof a company 1993 to Fiscal1994. that is not derivingopeating profits fiom its main 2.48 In Junc 1994,based on the report of the businessis appropriaocFor example,therc is a AssetDisposal Council, the Miisty of Trnspor- question as to whether thcir provisionof railway tationand JNRSC choseJR West to be the com- serves isobligatoryornot, that is,wheerornot pany lised in Fiscal 1994, while the scheduleand rheycandecidetodiscontinuesomeoftheirmoney- numberofstodcssold-offweretobestillxammind. losingrailway services. In Oaober 1994, one third of Japan Tobacco's 2.51 Anotherissuearoseconceningthenature shars wereoffered to the biddersand public,but a of the ManagementStabilizing Fund. The equity substandalportionofthesesharesremainedunsoldcapital of the three islandcompanies is relatively becausethe seingpricewas too high.On Novem- small, which makes the taeover easier in these ber 15, 1994 the Govenmcnt dccidedtde salesof cases,andtheamountoftheManagemcntStabiliz- JR West's shars should be postponea until Fiscal ing Fund is considerablylarger. For this rcason, 1995. The Govermnent considered that the fewrshve beenvoicedregardingtheacquisitionof absorbativecapacity of the stock maket for priva- the companiesfor the purpose of conrroll:ngthe tized-publiccompanies was not big enough. Ow- hugeamountoftheManagementStabilizingFund, ingtothisdecision,thecompletionofthesingoff in the casethat the stocksare soldoffwithout any proceduresfor the sharesof the three Honshu JRs restiction on stick-holdes. Consideringthe is- has been delayedstill furher. sues, the timing of the disposalof the three Isiand ShareS&Efir Nox-Himhu Rs JRs' stock is not dcearat present. 2.52 ConceringJR Freight,in addicionto the 2.49 The question of what to do regprdingthe need for maintaininga cercainlvel of profits in saleofthe renainingJRs'stock isyet to be decided. order to achieve listing on the exchange, most The three Island JRs (JR Holdaido, JR Kyushu, analystsagree that many questions remain unre- Japanese National RailwaysPrivatization Study II 19 solved,such as cost allocationwidt passengerJRs, ers. All this making stockdisposal problematical in the burden of new investnents, compeition with the near future. trudcs,restrictions in expandingbusiness, and oth-

Table/9-9:MarketConditionbeforeandin1992 TradingVolume TradingValue NikkeiStock Avg. (225) Avrage (million stocks per day) (billion Yper day) (V) 1988 1,021 1,025 27,039 1989 877 1,309 34,059 1990 484 717 29,437 1991 373 435 24,295 1992 Jan.-June 261 250 19,416 July 228 189 16,278 Aug. 311 251 15,790 Note:Trading Volumeand Trading Valueae basedon dieTokyo SwockExchanges fi section.

Tablei10-:Trends in ProfitandLossforthe ThreelslandJRs (bi/on yen) Hokkaido Shikoku Kyushu FY 1987 FY 1993 FY1987 FY 1993 FY 1987 FY 1993 Opauing Revenue 92 106 35 50 130 173 Oprating Loss - 54 -47 - 15 - 11 - 29 -27 Revenuefiom MSF 50 44 15 14 28 25 Currnt Pxofit - 2 1 1 2 2 1 II. JRs'Recent Performance and Newly ArisingIssues 3.1 TheperformanceofthenewlyacratedJRs ship. During the days of JNR, this public aspect has been remarkblc, although the growth of pas- had been emphasized.Afier th resuctring in sengersandtheprofirabilitybegan to sagnatefrom 1987. as a reacion to the past ecperiences,the 1992 owing to the prolongedrccession and other pendulum has swungto the oppositeereme and rasons. The environment surrounding each JR oeializedJRs are reiLcrantto underctk new di&rs, but dre improved responseto local necds lineconstruczion.Anotherissueisthepredicament and the bettr performance,whih is commonly of JR Frcight in the last few years. When tie observedin eachJRis delalythepositivesideofthe recessionset in and the rcdundant capacity of privatizaion. On the other hand, new issuesare existingtruck fleet emrged, JR Freih's lack of arising.Among them, one major issue is the ques- markcng abilityin th competitionagainst trucks tionofnewlineconsrmraion.Railroadbusinesshas became visible. Thc issuc is complicatd by the somepublicaspcsbynaurc, regardlessofowner- allegedcross subsidizaion cxisting both in train TablNI-: Opeang ResultsofJAs during the FirstSn Ye=sof Opeaon (iNAO r%,IWDntoi%)%Pa 1987 1988 1989 1990 1991 1992 1993 JR Hokkaido 3,920 4.457 4,373 4,631 4,800 4,9 4,800 - 13.7 -1.9 5.9 3.6 1.4 - 0.9 JR Eas 104,491 109,796 113,185 117,643 125,974 128,486 128,900 - 5.1 3.1 3.9 7.1 2.0 0.3 JR Centrl 41,148 45,085 46,302 51,051 52,110 51,200 51,000 - 9.6 2.7 10.3 2.0 -1.7 - 0.4 JR West 45,782 48,227 48,885 52,125 53,690 54,423 54,600 - 5.3 1.4 6.6 3.0 1.4 0.4 JR Shikoku 1,674 2,119 2,034 2,074 2,123 2,070 2,100 - 26.6 -4.0 2.0 2.4 -2.5 0.7 JR Kyushu 7,664 7,904 7,893 7,994 8,333 8,600 8,600 - 3.1 -0.1 1.3 4.2 3.2 -0.2 JR Passengers204,677 217,586 222,671 235,518 247,030 249,600 250,000 Tota 3.2 6.3 2.3 5.8 4.9 1.0 0.2 JR Freight 20,113 23,131 24,769 26,819 26,791 26,300 25,100 Tota 0.0 15.0 7.1 8.3 -0.1 -1.9 -45 Note: Figuresof de secondlinc showthe percencage duange from the pieviousyear. Source:Minisy ofTransporc.JRs. Japanese National RailwaysPrivatization Study II 21 and trudc transportation.Further complication railroadfrcight. In addition, containcrsof arisesif extrnalities such as pollutionand conges- new type owned by shippershave also con- tion are taken into account. Regulation by the tributedto thc trafficincreasc-, a 30 feet govcrnmcntof the privatizedJRs is another issue. container,a cool container, a small 2 ton This chapterdeals with thesc issucsas well as JRs' container, a 'car pack" (for automobiles), recentperfornance. and a largc tank container. pR's Recent Performance 3.5 Besidesthesc efforts made by thcJRs, two outside fiLaorscontributed to the growth of rail- Recent Developmext ofJRs' 9peration road passengers.One is thc inauguration of the 3.2 The growth of JRs' passengerssince the "Se Ohashi bridge between the main island of privatizationis remarkable,although the rate of Honshuand Shikokuin 1988and "Scikan" tunnel growthsloweddownin 1992andwasalmoststatic betwoen Honshu and Hokkaido in 1989, which in 1993. Freight transportationalso grew in the connected ticsc islands by railwaysfor the first period up to 1990, but has declinedsince 1991. time. The other is the fivorableeconomicenviron- 3.3 Sincetheprivatization oftheJNRinApril ment up to FY 1990. As a rcsult, the achicvement 1987, the JRs have rnade enormous efforts to has been impressivecompan-r to thosein theyears rationalizetheir operationand management.First, of the JNR and evcn to the PRs during the same they dramcally reduced their employeesat the perioa- time ofpuivatizationi(from 277,020cmployces to 3.6 In 1992, however, the rate of passenger 200,650) and then continuouslymade effortsto growth slowd down drasticallywith the burst of cut their labor forcesmainly in the railwayopera- 'bubble" economic growdL The JR Central suf- dons (from 164,671 employeesof the sevenJRs in feed mostamong the threc HonshupassengerJRs, FY 1987 to 136,081in FY 1992). as it relicsheavily on the busness passengersofthe 3.4 Secondly,JRs have reviewedtheir invest- builettrain,andregisteredthefirstnegativegrowth ment strategies,according to the maket charater- ofopration. In thisyearthegrowthofroralpassen- isoa in each regionof opation. During dhefinat grJRswas 1.0%.Theprolongedtrcessioncontin- yearsof the JNR total invnent had been re- uedin 1993andthreepassengerJRs(JRHokkaido, ducedto a minimallevl duetO the hugedeficit in JR Central and JR Kyushu)experienced slight opeons.Further,investmentdecisionshadbee declinesin passengergrowdt The growth of total made at the headquarterin Tokyo.The resultwas pangarJRs fel to only 0.2%. datiwmienrdecisionsdidnotrrelectlocalneeds. 3.7 The Japan Freight Railway Company, Sincethe privatization.each JR has seen respon- which registereda negativegrowth in three con- sible for its own investmentplanning secutve years,was most severelya&aced, as it had a. investmentin trafficcapacity and seice to competcagainst trudcswith excesscapacity in improvement(eg. fiequency,new type of dte lastfewyears, although their demandhad been coaches)in such metropolitanareas as To- above the capacay of existing flets during the kyo (JR East), NagoyaUR Central), and period of the "bubble economic growth. The Osaka (JR West) has seena great success siuation worsened in 1993 owing to the cargo comparedto that of the PRs (privaterail- unceintiscausedbyunusuallyfrequenttyphoons ways). in the southernpart of Japan and to the drastic b. acceleratedinvestment in trunk lines of the reductionof rice crop transportationcaused by the JR Shikokuand the Tokaido Shinkansen abnormallycold summer. has improved the competitivenss of the railwayagainst other modes of transport, Pi, wMor Caem gfp2r i.e. road and airway. 3.8 In line with opeating volumes, the rev- c railway'sdirect accessto an airport at enueof thejRs also increasedremarkably up to FY Chitose in Hokkaidohas cncouragedthe 1991 and then stgnated, becausefares have not total traffic demand, as well as strengthened beenchangedsince1987. Ontheotherhand,some the rway's competitvenessagainst road cost componentsof the JRs continuedto increase, transport. resultinginthe delineofboth operatingprofitand d. investmentin new type of rollingstocks for retainedearnings for the most oftheJRs duing the freight has encourageda modal shift to past three yars. 22 JapaneseNational RailwaysPrrnaization Study II 3.9 Along with the stagnant transportation for the JR Shikoku,and 1387.7 billionfor the JR demand for the JRs, severalfactors are responsible Kyushu.The profit arising from the Fund will be for the decline of their profit and earnings.These used to supplementtheir revenues. fators are discussedhereafter by eachgroup ofthe 3.11 The Law Concerning PassengerRailway JRs.For the three HonshuJRs, the purchaseofthe Companiesand the Japan Freight RailwayCom- Shinknsen lines in FY 1991 has had a significant pany (the Lawhereafter) provides the Fund in the negativeimpaa on their profit (althoughoperating form of debt owed to the three IldandJRs by the profits appeared to be increasing,current profits JNR Settlement Corporation for a redeemable decreased),as isshown by the iiusttive fgrcs in period of 0years(with a grace period of 2 yars) Table 1114. andasemiannualinstallmcntrepaymentplan(with 3.10 For the three islandpassgerJRs, delinc an interestrate of 7.3%). The rationalefor provid- inreveucsfomthcManagementStabilizingFund ingacne-tmefundraherthanprovidingasubsidy (the Fund herefter) has been one of the most eachyearistogivetheJRs managementincentives important factorsbehind their profit decine. The for efficincy,improvemen. = The fund setsa ceiling Fund provid for each company amounts to: amount ofcompcnsationfor the kss from thclines 16812 billionfor theJRHoldcaido,V208.2 billion which have to be in operationfrom the viewpoint

TabIAb Renu,Operatng Proftand Rened Earing of JRP (4Blon, %) FY 1987 1988 1989 1990 1991 1992 1993 JRHokkaido Rrvenue 92 94 100 105 106 105 106 Profit -54 -53 -53 49 -48 -47 -45 (Ratio) (-58.5) (-56.7) (-52.8) (46.7) (45.4) (-44.4) (42.8) Earnings 1 2 0 1 1 0 0 JR East Revcnue 1,566 1,664 1,736 1,852 1,950 1,981 1,974 Profit 296 323 281 292 355* 439* 417 (Ratio) (18.9) (19.4) (16.2) (15.8) (18.2) (22.2) (21.1) Eamings 27 41 57 58 57 57 57 JRCenutl Revenue 875 969 1,003 1,103 1,131 1,111 1,114 Profit 72 102 114 133 288* 405* 391 (Ratio) (8.2) (10.6) (113) (12.1) (25.4) (36.4) (35.1) Earnings 17 35 67 53 56 33 26 JRWest Revenue 763 807 834 892 917 923 938 Profit 71 79 91 122 128* 144" 137 (Ratio) (9.3) (9.8) (10.9) (13.7) (13.9) (15.6) (14.6) Eanings 2 5 26 30 36 32 30 JR Shikoku Revenue 35 44 44 48 51 51 50 Profit -15 -11 -12 -9 -9 -11 -11 (Ratio) (-42.4) (-24.7) (-26.4) (-18.1) (-17.8) (-21.5) (-22.6) Earings 0 2 4 4 4 2 1 JRKyushu Revenue 130 140 144 151 160 167 173 Profit -29 -29 -29 -29 -28 -27 -27 (Ratio) (-22.2) (-20.4) (-20.0) (-19.0) (-17.5) (-16.4) (-15.5) Earings I 1 4 1 2 2 0 JR Freigb RFvenue 173 183 192 205 215 216 206 Profit 11 11 10 11 7 6 2 (Ratio) (6.5) (5.8) (5.1) (5.5) (3.1) (2.6) (0.9) arningsr 2 3 3 3 1 0 -3 Notes: Operang Revenue:Revenue from railvay business (Eue rwnue and othere3ene and side-business.not indudedsubsidies Operatig ExpendueGlabor, energ,repair, mrisoelaneous expnses. t (not indcudedtrporae ax)and depredation (Ratio):Raio of OpertingPrfit to Oprtg Revenue * OpratingproficforjREa jRCentl andjRWescin1991 and 1992incrasedducto hc decras ofShinklnsenleasepaymenc (asis discusedbWow). Japanese National RailwaysPrivatization Study II 23 oftheserviceforthc"nationalminimum.kAndthis furtiheras the proportion of the Fund managedby ceiling amount has been recognizedas a kind of the three IslandJRs increases.The -Case in Table managemcnttarget by both managenent and em- I1-4 illustrateshow lag the impactof the interest ployees.The Fund has worked verywell as a man- rate for the profit of thesc three WsandJRs will be agemcnt incentivefor these dureccompanies until when al the Fund ismanaged by tic JRs with recendy. market mates. 3.12 Ncvcrthless since FY 1989, when the . . redemptionof the princpal began, dth three island JRs have taken thc responsibilityfor the manage- 3.13 The othe uncertaintyfor the JRs' man- mcnt of the Fund, in order to maintain ie princi- agement is the profitabilityof thc diversifiedbusi- pal and interestyield at a levelsomewhere around ness. Alrhough the diversified business has the annual rate of 7.3%. Thc general economic contributed to reducing the number of employes situation has since changed drastically in Japan, in the raihayoperationofthcJRs, dteprofitability and interest rates have declined significandyrthe of the businesshas neither increasednor reduced averagebank lending ratc has decied from 7.7% their total profit significandy.especially for the in FY 1990 to 7.3% in FY 1991, to 5.3% in FY three IslandJRs as is shown inTable HI-5. 1992, and firther to 4.1% in FY 1993. Thus the 3.14 Furthermore,itshouldbenotedthatmany revcnuefromtheFundfor thehrecompanieshas diversifiedbusinesses have alreadybeen scparated deaceasedand is expectedto declinesigni:andy from the JRs and nestedin to affiliatedor unaffili-

Table111-: hcrementa Cost for the ShinkansenPurchsed (IVBillion) JREast JR Central Operating Expenditurs 84.5 134 Depreciation 78.8 101 Fixd Property Tax, etc 5.7 33 Other Expenditures 198.4 314 Interst Pyment 198.4 314 Total Expendime (A) 282.9 448 Le= for Shinknsen (B) 210.3 416 Incamentad Cost (A - B) 72.6 32

TableW4: Opertng Lossand FundRevenue ( Bllion)

1987 1988 1989 1990 1991 1992 1993 Case* JR Hokkaido Operatng Loss -53.8 -53.3 -52.7 -49.0 -48.3 -46.8 - 45.4 -45.4 Revenue*" 49.8 49.8 49.6 49.0 49.0 46.9 44.4 28.0 Dif=exnce -4.0 -3.5 -3.1 0.0 0.7 0.1 -1.0 - 17.4 JR Shikoku OperatingLoss -14.9 -10.8 -11.6 -8.7 -9.0 -11.0 - 11.3 - 11.3 Revenue** 15.2 15.2 15.2 15.0 14.2 14.1 13.5 8.5 Differnc 0.3 4.4 3.6 6.3 5.2 3.1 2.2 -2.8 JR Kyushu Opeating Loss -28.8 -28.5 -28.7 -28.7 -28.1 -27.5 - 26.8 - 26.8 Revenue** 283 283 28.3 28.2 27.9 27.5 25.2 15.9 Differnc -0.5 -0.2 -0.4 -0.5 -0.2 0.0 - 1.6 - 10.9 Csasmesthesimation ofFY 1993 if dheinter of41% (actal einFY1993)isappliedforthemvenuefiom the Fund.AD ofthe Fundwill be rdmed to the duheIsdandiRs in FY1996. Revenuefrom dte Fund 24 JapaneseNational RailwaysPrivatization Study II atrdsubsidiarycompanies.Thusthefinancialstate- depreciationand retaincdcarnings (seethe Table mnentsof the JRs do not providea performanceof 111-7)in orderto improvetheirfinancialcondition. their divcrsifiedbusinesses as a whole. The only xceptionsarc the JR Centraland the JR Capttdand ExadE ztures F mg of Frcight,which havenot owneda largc proportion ath ERs of ther rail tack as assets and thus have nor tAcJRs registeredsufficiendy large amounts of deprecia- 3.15 As is shown in the tiable above, major tion. caelgoriesofJRs' invcstmentsduring the firstthree 3.17 The Lawprovides that an approvalby the tofouryearsare"maintenanceand rollingstocks" MinisterofTransportation is required for (a) new in order to make up the shortfall of these invest- stock issucs, (b) bond issues,and (c) long-term mnetsduring the finalycars as thcJNR. Sincethen, borrowingswith a maturityofmore than one year, the JRshave rapidlyincreased thcir investmentsin for all reJRs. Nevereless the Lawdoes not allow 'improvement of infriastructure,"for examplethe t three Island JRs to issue bonds, and instead improvement of celctricfacilities, advanced ATS provides the Management StabilizingFund to (AutomaticTrain Stop),ATC (AutomaticTrain strengthentheir financialbasis. Control) and CTC (Ccntralized Traffic Con- 3.18 The Lawprovides some favorsto thcJRs. trol), and railwaytechnology developmcnt. This Until the CommercialLaw was revised(effectve trend clearly shows that the JRs have changed from 1993), a provisionin the Lawhad alloweda their emphasis in investment from currently higher ciling for the amount ofbond issuesby the urgent projects to investmentswith a long-term three mainland passengerJRs and the JR Freight; perspective. dteseJRs can issuebonds up to ten times amount 3.16 Almost all the JRs have resutainedtheir oftdenetcapital valueofthclastavailablebalance investmentamount to the amount of theirdepre- shect(in gcnral, the amount of bonds is restricted ciation, or to the internal sources of funds, ie., uptothenetcapitalvalue).ThcTawalsostipulates

Table1-: Revenueand Profit of WiversilledBusiness OiBillon, %) 1987 1988 1989 1990 1991 1992 1993 JR HokkaidoRevenue 19.4 12.9 18.9 193 16.9 14.9 15.7 Profit -0.2 -1.8 -2.0 -0.7 -0.5 03 0.2 Ratio -1.3 -13.9 -10.4 -3.4 -2.9 1.9 I1 JR East Revenue 30.6 50.9 62.7 69.7 74.1 79.7 81.1 Profit 8.3 7.1 3.8 8.1 9.3 7.6 5.8 Ratio 27.1 14.0 6.0 11.6 12.5 9.5 7.1 JR Cntral Revenue 63 2.8 3.7 4.0 4.3 5.2 8.7 Profit 0.5 1.0 1.3 0.2 0.6 1.5 2.3 Ratio 7.6 35.5 35.5 5.9 13.8 28.4 26.4 JR West Revenue 16.2 12.6 11.2 12.7 15.4 16.5 23.3 Profit 0.7 0.8 2.3 4.0 6.5 7.2 8.1 Ratio 4.4 6.1 20.4 31.6 42.0 43.7 34.7 JRShikoku Revenue 4.6 3.1 2.9 3.1 8.0 9.5 7.8 Profit -0.9 -1.0 -1.6 -1.6 -4.1 -4.2 - 3.0 Rato -19.5 -32.2 -55.7 -50.9 -50.8 -43.6 - 27.4 JR. Kyushu Revenue 3.2 3.7 14.2 19.3 23.9 23.3 28.3 Profit -0.7 -0.8 -5.5 -7.8 -7.6 -6.7 - 5.4 Ratio -23.0 -22.7 -38.9 40.5 -31.9 -28.5 - 18.9 Tota Revenue 80.3 85.9 113.5 128.3 142.7 149.1 164.9 Profit 7.6 5.2 -1.7 2.3 4.2 5.7 8.0 Ratio 9.4 6.1 -1.5 1.8 2.9 3.8 4.9 Note Profit is the nurgm only afierdirct mast.and indiudessadi indre cmsas intrest payment had reprseus profitor loss over revenuc. JapaneseNational Raaways Pivatization Study II 25

Table 6: CapftalExpenditures by the JRs (fBIIIion) FY 1987 1988 1989 1990 1991 1992(Plan) 1993 (Plan) Maintenance 74 86 103 107 150 178 157 Improvement 22 41 93 133 153 211 211 CapacirExpansion 34 48 49 63 103 123 92 RolingStock 65 163 198 208 172 180 156 Others 15 14 16 18 20 22 24 Tota 208 353 459 528 598 714 640 *Ihis expendie doesnor indudethe pucdus of ShinkansnUnes.

TableE-7: Investment Compred to Oepreciaffonby theJRS (1987 - 1991) ( B1710ion) InteralFuncs Investment Tota Depreciaion Retained Earnings JR Hoklkaido 87 92 88 4 JR East 788 1,390 1.183 208 JR Central 702 598 405 193 JR West 309 558 476 82 JRShikoku 53 58 45 13 JR Kyushu 103 127 118 8 JR Freight 105 74 63 11 Total 2,147 2,898 2,378 520

Tablel-8: LongTem Enl Fino cingby SomeJRs (4Bkon) 1987 1988 1989 1990 1991 1992 Total JR East Bond" 106 97 96 94 62 120 575 Loan 42 342 461 35 46 81 1,007 ToMa 148 439 557 129 108 201 1,582

JR Centra Bondc' 19 20 19 20 18 20 116 Loan 16 48 83 38 58 73 316 Tota 35 68 102 58 76 93 432

JRWest Bond** 39 35 36 39 15 16 180 Loan 18 50 62 20 46 49 245 TotM 56 85 98 59 61 65 424

JRFreight Bond*** 2 5 8 9 12 0 36 Loan I 1 4 3 19 16 44 Tota 3 6 12 11 31 16 79 "Bond isu incdudebonds guaateed and subscribedby dueGovenme Bond issue in FY 1992 is a forign madke bond, and aD the odtes are domestic bonds guaraeed by the Goverment. ADthe bonds ate guaantd by the Govrnment 26 Japanese National RailwaysPrivatization Study II that up to the fifih year of operation,the Govcrn- brought considerablebenefits cothe Japanesetax- ment can providc a guarantee for the JRs' bond paying public in subsequentyears. issuesup to the amount approvedby the Diet. Efforts and ManagerialProblems ofjRs Financid Burden on Government ThbreeHombu IRsf 3.19 Thc financialburden on the government concerningthe JNR and the JRs has been drasti- JR callyreduced since the privatizationof the JNR, 3.22 Badckdby the largcstmarketo fthe Tokyo and the JRs have paid more taxes than subsidies metropolitanarea,theJR Easthas thelargestoper- received.Table III-9 showsthe amountofsubsidies ating basis among the JRs. The revenuc from the receivedfrom the central goverment, and the Tokyo metropolimanarea reachesabout half of the amounrofcaxs and contibutions paidbycheJNR companyIsrevenue from railwaybusiness. Of the andsucceding institutionsto the centraland local remaning half, about half comes from the two govenmuents. Shinasenline!, Thus the primary stratg of the 3.20 Bcefreprivatzarion, the amountofsubsi- company'sinve cmcntis the improvementoftrans- diespaid to the JNR exceededthe amount of taxes portation capacityand servicesin the Tokyo met- and conttibutions received from the JNR (as a ropolitan ara. For exuampl, the company has public corporation, the JNR was not required to introduced double-dckers and increased the nur'- payfdral maxes).Theamountofannualnetfinan- bers of trains with air conditioning. ITe second cal burden to the government rched almost priority of its invesment is the rational use of the !600 billion. eising trunk lines and Shikansens. A good ex- 3.21 Foilowingthe privatizaion of the JNR, ample is the Yamapta Shinkansenwhich directly hower, the JR companieswere required to pay connectsthe existng trunk lincwith a Shinkas Iedel taxes such as the corporate tax, and the to acceleratethe speedof the train. And the com- amount of such taxes paid was quite considerablc, pany has activelypromoted the corporateimagc as reflccingthe favorablemanagement conditions of a prvate railwaycompany. Bcsides these efforts by the JRs. In addition, the amount of subsidiesha., the company,the strong economicgrowth during decreased.As a result, the amount of rae came tO tde initialfiveyears ofits operationhas contribured exceedthat of subsidiespaid to the JR companies, to the company's impressiveperfirmancr. This implyingthat the plivateation of JNR has de- allowedthe company rtobe succsfuly listedto thEc creasedthefinancialbuxdento thegovemnmentand stock marketin Scptember1993.

TableM1-: Taxes and Subsidies (I' Bitlion) FiscalYear Taxes and Contributions Subsidies Balance UnerJNR 1982 35.2 729.4 -694.2 1983 38.3 701.8 -663.5 1984 46.3 647.4 - 601.1 1985 47.8 600.1 - 552.3 1986 50.1 377.6 - 327.5 UndCjR 1987 220.7 193.7 27.0 1988 245.7 211.0 34.7 1989 230.1 633.1 - 403.0 1990 299.4 157.1 142.3 1991 444.3 108.2 336.1 Not: This tble alsoappeas m ChapterIL For axs contlmbtionsand subsidicsfollowing dhe privatiion ofJNR. figuvrs reprcsema totalof thosefrom the JRs,JNRSC.the Shinkanseni:1lding Cosponion, and the RailwaDiclopmaent Fund. * Beuse tdeJNRSCcaries liabilities in additionto dhse ofJNR.some subsidies to thcJNRSCdid not exist in theJNRyer In 1989the amountof subsidiespaid by the govenmentinreased rapidly due m a specialmeasure concening pensions. Sourc=Ministay of Transport Japanese National RailwaysPrivatization Study II 27 3.23 Ncverthelessthe financial burden taken 42.8 lan whkh is currendy under construction overfrom the JNR and addictionalburden charged between Otsuki and Kofu. from the purchaseof the Shinkansensrenains the 3.26 The other strategicinvestment which has companysmajorconcern.Tecompany'snerworth beenimplementedcduringtheinitialseveralyersaof ratio declinedto 8.1 percentin March 1993 after it its operation,is improvementof the railwaycapac- had been improvedto 12.8 percentin March1990. ity and servicesin the Nagoyametropolitan area, And fixedliabilities amount to about threetimes its which is the only metropolitanwithin the operat- turnover.Thus the companyhas been limiting its ing area of the company.The company has intro- investmentamountto the intemalsourcesoffiinds, duced new type of trains which accommodate in orderto strengthenits financialcondition.With commuting passengersand has improved railway the listing on the stock market, the company ex- infrastructure which provides frequent services. pects to expandits fund raisingcapability. Nvcrthdess,Nagoyacityhasbeenprinarilydomi- 3.24 Theotherconcernisregulation,especially nated by road transportationand also well deCVe- the tariffsystem.In order to secureasound operat- oped with the local railwaynetwork by the PRs. ing profit and to respond to a varietyof denands, Thus the demand for the exsting JR lines seemsto flexibilityof tariff systemis a crucial factor for the be saturaed, or to increaseonly in line with popu- companys management.Now that it has been lation growth along the lines,after arapid increase listedto the stock narket, the companyadvocates since the privaizarion of the company. much morc flexibletariff regulation than befiore 3.27 As is the case with JR East, the low net Laudy,the diversifiedbusiness is onc of the impor- worth ratio is another concern for the company tant factors to strcngthen the company's profit fromafinancialaspect, especiallyafter thepurchase basis, althoughthe current levelof revenueis only of the Tokaido Shinkanse Fixed liabilkieshave marginalpart of the total turnover of the com- icreased to five timesthe companys tumover in pany. March 1993. Neverdhelssthe carrentprofit ratio 'ARCemral was as high as 10.4 percent in FY 1991 and still renmains6.1 percentin FY 1992which isquite high 3.25 The principal businessof the company is comparedto thoseof the PRs. Basedon this solid the Tokaido Shinkansen whkih has the largest profitability,the company hopesto be listedto the single-linedemand in Japan m terms of passenger- stock marketas soon as possible. km. second only to the Tokaido line which runs paralletotheShinkansen.Therevnucshareofthe JR We TokaidoShinkansenoutofthetotalturnowverofthe 3.28 Two major sourcesof revenue for the JR company isas high as 87 percent.Thus the crucial We-t is the Shinkansenand Osaka metropolitan iSsus for the company are (a) maintenance,im- area,whih is similarto the JR East.Nevertheless, provement,and capacitycxpansion of the existing the trafficdensity oftheJR in the Osaka metropoli- Tokaido Shinkansenand (b) investment for the tan ismuch less than that in the Tokyo metropoli- newmChuo Shinkausenline which is expeacd to tanarca, mainlybecauseofthelessequippcdrailway employ"Linear Motor technology.This tcchnol- network in the region. Although the total route ogy, using a superconductingmagnet, has been lengthof the JR West in the Osaka metropolitan tested by the "MiyazakiTesting ULneof 7 km in arcais7OpcrcntofthatofthcJR EastintheTokyo Kyushu,ahieving 517 km per hour, and will be metropolitanarea, the track runs around in the further tested in the "Yananashi Testing Line" of regionwith a less densetraffic demand. Asa result,

Table111-10: Rout Lengthand Trafficof JR East(FY 1992) (kbn,million passenger-kn million) Route Length (A) Passenger-kn (B) Density (B/A) Shinkansen 839 16,244 19.4 Tokyo Met. Area 1,097 76,755 70.0 Otier Lines 5,566 35,487 6.4 Total 7,502 128,486 17.1 Soure: JR Eas. KabushikiT7dsk-de Toroek Mokemi Sho()fV*e=ff). 1993 28 Japanese National RailwaysPrivatization Study II the demand (density)for theJR West in the Osaka istrationwhich is similar to the railwayadministra- metropolitanarea is only half of that for the PRs. tion officeof theJR Hokkaidoand theJR Kyushu. This is a good contrast to the JR East in the Tokyo The idea is not to increaseprofit ability, but to metropolitanarca which sharcs the dense railway reduce the loss,mainly through (a) laborsavingby market almost equally with the PRs. The other integratingthelaborfunctionsand(b)icductionof characteristicsofthe JR West is itslarge share ofthe maintenancecost by applyingdciferent standards locallines, comparedto the other Honshu JRs. In of maintenanceamnong ines. addition, its trunk lines have a relativelysmall Three bld Rs trafficdemand. In other words, the company is a typical orginization which stil has an internal JR Hokkaido 3crosssubsidizaton.t 3.31 Located in the most sparselypopulated 3.29 Strategy on the positiveside for the JR areaofnordhernislandinJapan,theJRHokkaido West is thus foacsedmainly on the speedup of the has an inherentlydifficult managementenviron- Shinksen andthe developmnentof the urban mcnt both for railway business and for railvay networkin th Osaka metropolitanarea. In order rdated businessand the viability of the railroad to compete against the airway,devldopment of a sectoritsdfhas been repeatedlyquestioned. At the new train for the Shinhan3en namedhuWIN350s timc ofprivatizationsome unprofitabl lines rwere which has acieved a 350 km per hour is crrently abolishcd,or separatedfrom JR Hokkaido, but it being undertake As for the developmentof the alsohad to acquiremany unprofitabiclines which urban networkin the Osakahmetropolitanarea, the could not be abolished fiom the viewpoint of company has been constmuctingsvcral lines and "socialminimum". Thus it has the weakestfinan- expandingthe linecapacity by doubletracking and cialsructure amongthe sixpassengerJRsand relics elctrificaton. Residential development in the heavilyonsuchfctorsastheM nntStabiiz- Osaka metropolitanarea has beenspreading out to ing Fund and avorabl tax treatment. theregion where the PRsdo not havean perating 3.32 Neverteless the companyhas achieed a iniruccure. Good examplesare, (a) construc- etteroperatingrcsult during thefirstsixyears than tion of a new line calld "Kaafutku line" which wvasex opecte.Wth the suclcssof the newairport connrmcstheeast gate ofthe Osakacity a machi accessline which connected ChieoseAirport and and Kyobashi) and tethe westlos the Osakpacty -Sapporo(the largestcity)andwhih ovaiwhded (Amagasald),and (b) the double trackingof part of the bus service,the companyis now confidentthat the Fukucyama line wher the spreadingout of the railroadsector continues to be viablein ain the residentialarea is veryactve. trunk lines originatingfrom Sapporo. The airport 3.30 The other side of the strategy for the accessline isone exampleof the successfilaspecs companyisrationalizationofchelocallinemanage- ofprivatization.With the cxpansionof air passen- ment. The JRs acknowledgctheir obligation to gers, the ter i:uinalof ChitoseAirport was planned provide railwayservices on local lines that are not to move to an adjacent area.When the initialpan profitable.The tariffon theselocal lines is allowed of newairportwas disdosed in thcearly 1980s, to beonly 10percenthigherthan the normaltarifE the rewacionofwas itheearldisclosedm 198 ow. Thus thereis no other choicefor te JRs than to They argue,' that air passengerswould use taxiszo minme the operatingloss fiom these locallines. and firomthd nearbytrain stationbecause the new Thus the companyhas establishedrailway admin- terminalwa .oo fiLrtowalk The new managemenr

TableIllI: RouteLngth andTraffic of JR West(FY 1992) (kn, millkn pssengerkm, million) Route Length (A) Passenger-km(B) Density (B/A) Shinhansen 645 16,161 25.1 Osaka Met Area 765 25,419 33.2 OtherTrunkLincs 2,046 10,932 5.3 LocalLines 1,603 1,910 1.2 Total 5,059 54,422 10.8 Source JR WcslJR Nisii hn Iha Yeram(Profik efIR Wat). 1993 JapaneseNational Railways Privatization Study II 29 ofJRHokcaido,whichwasnaturallykeenonlocal its investmcnt has been effectivelyallocated to demand,realized that all air passengerswould use those lines with comparativelydense traffic de- bus scrviceboth timewiseand moncywiseif the mand,namelythelineswhichconnectsAsahikawa, altcernativewasthecombinadonoftrainandtaxi.In Hakodate and Kushirowith Sapporo. the firstyears aftcr the privatization,the construc- 3.34 Theotheraspectismanagcemntcffciency tion oftheairport accesslineand the investmentin improvementin the other local lines,which con- newroilingstockforthislinebecamethefirstpriority tinue to lose money. In return for the existenceof ofJR Hokkaidoand the line turnedout to be a grcat the ManagemcntStabilization Fund, thc company su osscncethenewairportwasopenedinJuly1992. has the responsibilityof passengertrain operation 3.33 The company'sbusiness strategy has in and the idea is to rcspond to local nceds and to princple two aspects.The frs onc isactive invest- minimize the loss.One exampleis cstablishment of ment and competitivenessimprovement for the a railway administration office for independent major lines (trunk linesand an airport accss line) locallines, as the JR Wcst and the JR Kyushudo, that originatedfrom Sapporo.About ninety per- whichcan be describedas 'AutonomousMarginal cent of the total passenger-kmof the companyis lineManagement.Theadvantagesofthisscheme transportedby these major trunk lines. Since the are, (a)more flexibletime schedulingreflecting the investmentfind resourceofthe company has becn local needs and (b) labor saving by integrating in principlelimited to the internalsources offunds, variousfunctions necessaryfor train operation.

Chartlif-I Hokkaido RailwayCompany /.,..

, _ gb Ll I~~~ftbLm 30 JapaneseNational RailwaysPrivatization Study II 3.35 Current major concernsof the company and (c) improvementofprofitabilitryofthedivezsi- are, (a) declining revenuefiom the Management fled businesses. StabilizingFund (mentionedin the precedingsec- 3.38 The company'sefforts for the speed-upto tion), (b) increase of tax burden at the time of competeagainsthighwaysareremarkableandmust expirationof the specialtax treatmentin FY 1997, be noted as the visiblepositive sidc of the privatiza- and(c) heavyinvestsnentrequirement forthe main- tion. The electrificationof Yosan line, one of the tenanceofrail trdcin the'Seilan tunnel 10 to 15 trunk line which connects Takamatsu and yearsfrom now.Thcfinancial impact ofthe second Matsuyama,has been completedin March 1993. factor is estimatedto amount to Y7.4billion, In October1993, double-trackinghas been com- whereas thLecompanys net carnings in FY 1992 pletedberweenSakaideaxidMarugame,partofthe amountcdtoYO.3 billion. Expectedinvestment for Yosanline and the part wherethe trafficdensity is the d!ird ctorexceeds 1100 billion, whereastotal reativly high. The Dosan Line, the other trnk investment during the first five years amounts to line, whichconnects Takamatsu and Kouchi,runs Y87billion . througha mountainous area wih winding rail JR Sbikku tuades.Here, fforts haveben concentratedon the introductionof new tilting trains,which tilts at the 3.36 Atthetimwofprivadzaion,thiJRShikoku curve to make each car more resistantto the cen- bad the least developedrailroad infrastructure in tiugal force, dius enablingthe rminto maintain terms of sharesof electificationand double track- the speed.The other effortsin this Dosan Line is ing. In addition, hamperedby the mountains, the the snrigning of one track in each station.The routes ofthe linesare forced tohave alot ofcurves. Dosan Line is still single-tracked,but in each Suinc then the company has promoted the tain station it becomes doublc-traed, naldng each speedbothbyecrifcationandbyrintroductionof tracdcauved to both sides of a platfomL Until newtrains thatcan tiltoncurveswhilemaintaining recendy,when an expresstrain passedeach station, speed (tilting trains). The share of the electrified it had tOreduce the sped unnecessarilydue to the route length of the compay has been incrased curveofthetrackNowonectackhasbeenchanged fromon6.4 percet inApril 1987 to 27.5 percent to be saightc at eachstation and an expresscan pass inMarch 1993. In the matnie, the -Seo Ohashi" throughwith highe speed. btidgec,whichconnects theShilokuislandwiththe 3.39 Asaresultofthesecfiorts,passegersofthe Honshu (mainland) by rail and road, has becn company d not decreasein l993 and has shown inaugurated in April 1988 ar.-l contributed to a aslightincreascofO.7%inspiteoftheinauguration significant increaseof passengers,27 percent in- of two direcdy competing highways.The morale crease of passenger-mnin FY 1988. As for dte behind these efforts has beemsupported by the efficincy improvementof the local line manage- senseof crisisthat the companycan not relyon the ment, operation without a conductor has been nation anymorc after privauzation.Both the man- enlcouragedand minorstations h;ve ben dosed on agementand the labor realizedthe localsituation the weekend. As a resu;k supported also by the and has madeeffors for the survivalofthe railroad revenuefrom the ManagementStabilizing Fund, the sector in ShikohL companybasregisteredaconstantprofitsince1987. 3.40 The company on its own has the highest 3.37 Nevcrteclessthe major challengefor the shareof revenuesfrom diversifiedbusiness among company is competitionfrom road trafficsince FY the pasenger JRs. Althoughthesc divcrsified busi- 1992, owing to the inaugurationof the two high- nesseshave been making losses,the company has ways which covers major traffic market in the been able to realize wnergyeffecs with the main Shikoku island, betweenTakamatsu and Kouchi, businessof railway,for eammpleresidential devel- and between Takamatsu and Saijo, which run opment (as the PRs do). In addition,by operaing almosr parllel to the company's two trunk lines highwaybuses in cerain seleccedroutes, the com- (SeeChart I1I-2).As a result,and alsoowing to the panybassoughtwaysto compensatefor the decline recession,the passenger-kmin FY 1992 has been oftherailhapassegsshiifingtotheroadtuaffic. decreasedby 2.5 perent. Requiredstrategies ofthe company are; (a) furtierspeed-up investment for JRI(ah the trunk lines in competitionwith hiways, (b) 3.41 Among the three smalllsland JRs, the JR firther efficiencyimprovement for dte local lines, Kyushu has the best marker for the railwaybusi- JapaneseNational RailwaysPuivaization Study U 31 ness.The islandhas twolarge cities with a popula- 10% and frequencyofutain by40% from 1987 to tion of over one million, and also has several 1992 to servebeter the commuring demand.Al- medium sizecities. EspeCallyin Fuklcokametro- most onc fifth of 1,500rolling stock unit are newly politan area,the populationhas been grwing and built, andatmre improvedaccommodation.Thus the residentialareahas been sprcadingout. In order the compny has achieved a stable inces of to utilize thesc advantageof the madret,the com- passengerdemand. The number of passengershas pany has incrcasedboth the numberofstations by inceased at a greaterrate than passenger-kin,be-

Chad1-2

ShikokuRailBway Company Md

HIighwayNetwork miSbikoku

sar

m) tH \t/ 1n - n~~~~0===Ita"vg cr Cntmct6ng 32 Japanese National RailwaysPrivatizaion StudyII causeshort-distance travel passengcrs in mcropoli- the urban lines, the companywill further improve tan arcas are growing. the railway servicc for commuting passengers,cap- 3.42 LikeotherJRs,JRKyushu'sbusincssstrat- turing the spreadingout effect of the residendal egy has two aspects.One is thc concentrationof area investmenton stcategicallyimportant lines and on F lines wherc railroad can be viablc with cfforts. These currentlyconsists of new MiyazakiAirport RecentPeiformaceofJRFreg: accessline, speedup betweenFukuoka and Oica 3.46 The JR Freight is the only railwayfright with new rolling stock in each case, and new cmpany with a nationwide network in Japan. Kyushu Shinkansen.For instance,it will cake60 Contrary to the general pessimisticforecast for minutes from Nobeoka to Miyazakiairport with railwayfreight, the company has registereda trc- the inauguration of airport access linc and the mcndousgrowthduringthe initialfouryearsofits completionof new rollingstock, 13 mninutesfaster opton, and has almost alwaysexceded the than the prescnt journey time. In the case of the projectedfigures by the Govemment The reans airport accessline, using funds from central and for the revvl of the railwayfreight can be attrib- localgovernmnents, the companyhas been able to uted to followingfiacors; providc V1 billion of the tomalY6 billion projcCt a) the strong fight trafficdemand in general. cost. Thus in theselines the companyis preparing due to the raid ecnomic growth to compee with roadDs. b) the modal shift from the road sector, be- 3.43 The other factor is the effidencyimprove- causeof the shorcagcof the truck driecrs ment of money-losinglocal lines. In order to im- and road trafficcongestion prove the localline management,the companyhas c) the completionof the Seto Ohashi" bridge introduced railway adminiscration as the JR and the "Seikan"tunnel which connect the Hokkaidoand thcJR Wcstdo. labor functionsare Honshu (mainland)and Shikokuand integratedto reduce labor forces,and an operation Holdaido islandsby rail without a conductor has becn encouraged.Thus d) the restraint from the tariff increase,despite the company has registered an almost conscant the truic tariff increase profit during the first sixyeas of opcration,sUp- e) serviceimprovement, for exampl incse ported by the revenucfrom thc ManagementSta- of the number of despawches,introduction bilizing Fund. of varioustypes of containersand wagons 3.44 Ncverthedess,rejgardingrheimportantlines, 3.47 Nevertheless,asthcjapancseeconomyhas competitionagainst bot highwaysand airways, fllen into recession,railway fight demand has cspeciallyhighways is expectedto bc formidablein againdelined by 1.9%in FYI992 and the profitof the Kyushuisland. For example,currendy bwcen the companydecreasedto barelyzero,even though Fukuokaand Oita it takes 147minutes by train, 30 the companyhasbeen subsidized by the 'avoidable minutes faster than by highwaybus. However, cost"principle(describedinthesectionfiom3.104 when the highwaybetween Fulcuokaand Oita is on). The situation worsenedin 1993 due to the completedin 1995,buses will connecttwo cities in decreascof cargocscaused by unusuallyfrequent 140 minutes. In order to compete against high- typhoonsin the summer and by the sharp drop of ways, the company is improving the speed oftrain, governmentally purdcased rice crps. resulting in and witi the compltion of new roiling stockS,it the loss of 3 billion yen in hkisyear. will cake133 minutesbetween two cities. 3.45 Thus the company has a plan to firther MarhcinranComon improve the speed of trains on the intercity lines, 3.48 Themostimporcancissuesarcthemarlcet- includingthedevclopmentofthenewShinsn ing and the competition from the road traffic. The firststage of the newShinkansenis plannedto Whilc the key strategyfor the company is shifting be completedin 2001 in southernKyushu. When the rail -ayfreight businessfrom bulk cargoesto dhisShinkansenbetweenYatsushiroandKagoshima containers, the company does not have a direct is inaugurated, it will deczeasethe time between contact with dicnts. Instead, it relies on a single Fukuokaand Kagoshimafrom 250 minutesto 150 frcight forwarder,ie., Nittsu (the biggest -4ight minutes.Thcn, JR Kyushuchinks char train willbe transportation company),for pickup and deley more advantageousthan air travel.In addition,in services,and for marketingas welLBesides, Nitsu JapaneseNational RailwvaysPrivzation Stu I 33

Chat111-3

Kyushu Railway Company

-i

hkiy h aS~W--

LwJ 34 JapaneseNational RailwaysPrivatization Study II itsCtf is facing a severecompetition from other lcased such facilitics as containers and forklifis. forwardersby road taffic. As a rcsult, fiom the Thus it isestimatedthatthcrealinvestrcentduring balanceofbargainingpower,thecompanyisforced the first five years should have been Y120to 130 to providea largediscount. Thus it is desirablefor billion rather than the actualfigure of V105 billion the companyto scrcngthenmarketing and to select , if ctheleased facilities are taken into account. and cany out only profitablc business.Another 3.51 The company leasesalmost all of the rail issue is to reducethe labor forcesto an extent that track fiom the passengerJRs (the leased crack is compatibkcwith profitableoperations. readces 9,978 km, whereas the company's own In vestnme RequireatRolling Stock track reached only 76 km at the end of March 1993). The company and the passengerJRs have 3.49 Althoughthe company'sfinancial base is reachedagreement that thc passengerJRs should weak, it is rquired to invcst a largc amount for maketrackavailableforthecompany(acthcsecond rollingstockandinfirstrucure. DuringFY1989 to priority together with supcr Cxprcss and Cexprcss 1992, the total inavcstmcntby the company trains as thc samc order, only second to the bullet amounted tO Y123 bilion, whereasthe intral trains and commuting tminsin the metropolitan sourc of funds amounted to only 60 billion, of areas as thc first priority.) Ncvertheless,as the which 53 billion camc from depreciation.This demandfor the passengcrJRshas incrcasedsignifi- shortfaUin thc internal fimd sourceresulted in an candy during the first severalyears after privauiza- increaseofthelong-tcrmdebtroutstandingamount tion, it has becomcdifficult for the passengerJRsw of bonds and long-termloans has inceased from make a track, especiallyof the Tokaido linc, suffi- Y78billion in March 1989 to Y93bilion in March cientlyavailable for the freight taffic. 1993. On the other hand, since the JNR hardly 3.52 Thus the capacityincrease of the Tokaido investedin the rollingstock during its lastsevcral linc, which has the largCstdemand (about thirty ycars (and actually the investmcnt in the freight percent) for the raiway feight, has become an seEorwas allocatedthe least priorityin the last ten importantste& forthcompany.Thccompany yearsofJNR), investmentin rollingstodc has been has a plan to invest Y30 billionfor this project, of and continues to bc an urgent priority for the whichIllbillionwillbeinvestedforthefirststage company. duringFY1993 to 1996.Capaciryexpansion ofthe 3.50 Inorderto alleviateitsfinancialburdenfor powuersubstaionscostsabouthalfofthe 11 billion, roling srodc investment,the company has pardy and the remaining amount will bc invested for

Table11-12: Capital Expenditures by the JR Freight (I Billion) FY 1987 1988 1989 1990 1991 1992 1993(Plan) Maintenance 0.9 1.6 3.2 7.3 20.5 20.3 6.7 Improvement 1.5 2.0 2.0 3.9 4.6 6.2 11.6 RollingSwock 2.0 11.9 15.0 17.4 10.7 11.8 12.2 Total 4.4 15.5 20.2 28.6 36.0 38.4 30.5 IntcrnalFund 12.1 16.7 15.5 16.1 13.9 14.1 n.a. Depreciation 10.3 13.6 12.6 13.3 133 13.9 n.2. RetainedEarnings 1.8 3.1 2.9 2.8 0.6 0.2 n.a. *'Maintenance figurasin FY 1991and 1992 indudeinvesment fiorwarehouseswhich amounm to mowrhdunV1ObiOionannuaily.

Tabl W13 ContainerInvestment Purchase and Lease (I Billion) FY 1987 1988 1989 1990 1991 Total Purchase 3.0 7.7 53 3.0 0 19.0 Iease 0 0 5.0 9.2 7.0 21.2 Totl 3.0 7.7 103 12.2 7.0 40.2 Japanese National RailwaysPrivatization Study II 35 expansionof the freight terminal and of the side- in the followingsection 3.60) to operatethe aban- track This project has been authorized by the donedpartoftheline,whichthecompanylkasesby RailwayDevelopmentFundand thusthe financing payingthe current "avoidablecost." that the companyisresponsible forwill coverabout New L'meConswuction and Conflictsof Public two thirdsof t totalinvement cost.The reain- T der will be financedfrom the Fund, byan interest- frec loan, from the 'Fiscal Investmentand Loan Grouing Corq%a efIntffat Programs,"and from bond issuesby the JRCPC 3.56 Sincc it usually costs a huge amount of (japan RailwayConstuctionPublicCooperation). moneyto construct a new railwaylinc, it is impos- 3.53 The first stage of the investmentwull in- sibletotransferthewhoecconstructioncostdirecdy crease the number of the 1.300 confteight trs to the tariff of the new line. In addition, it is also Bythe secondstage investmcnt, the tain speedwill dfficult to get consensusto tansfer the construc- be improvedfurthcr. And la.ty, by the third stagc tion costofa specfic line to the generaltariffof the investnent, a 1,600ton freight trin willbe intro- railwaysrvic. On the contray, the railwaydevel- duced. opment providesextcrnal economics for the real 3.54 Anotherpotential projects for the capacity csmtebusimssaround the railway,rcsideats around increaseis the 'Nanpo Kamotsu Sen (Southern the railwaystations, and the local municipalgov- FreightLine)' in the Nagoyaarea. Its consrucon emment in the form of increaseof tax revenucs. started in 1967 is yet to be finishedwith comple- Thus thereahvays exists a conflictofnterestamong ron ofabout 95 percent;Ohfi - Inaawa of2l kn. those who bear the cost and those who receive The rest of the most congesed area for Tokaido benefit.In other words, someschemes for sharing Line, ie., Tsurumi - Odawara (about 60 Ian) in the development profit, or arrngement for the Tokyo area and Ibaragi - Nishiakashi(about 70 iteraliation of external coonomiesshould be km) in Osakaarea, has beenequipped as quadruple mamisd tracling. Thus the Nagoyaarca has become the botdeneckfor the capacityincrease of the Tokaido baerndan ofFvernd Economies lie. But the additional invemnt cost is esd- 3.57 Theseschemes are dassified in principle nuted to V10-20billion that the companycannot into three ypes.The firstone is an inenalion casilybcar. of exernal economieswithin a railwaycompany Re,ining theNatomiwide Freigh Raiwa Network vvhi developsa new linc togetherwith land The second one is a direct trnsfcr of the external 3.55 Anothe seriousproblem in futureth e is economicsfrom the beneficiaryto the developerof rlated to thc NewShinkansen Lince pmen a new line. And the las one is an indirect tcransfr Plan discussed below.Wh all the lan of t xteral conomies by imposingan extra tax Shlinkcansenlines are built,part of the currentlocal for the lbeneflt. trunklinesmaybeabandonedbythepasengrJRs. th5 file be i . Thuse cmpany may hv to prase and mented in the U.S., aimirvgprincipaly to reduce operate that specificpt of the abandonedtrunk financialburden for the localgovcrnments who are lines and to bear he whole cost of the capital, the important investors for urban railways,espe- maintenance,and opaation in order to maintain cially aftcr the 'Tax Reform Act of 1986"; (a) its unifiednationwide track system.But it incursa Increaseof sale tx in the concernd region; (b) huge amount of cost for the companyto own the Establishmncnof "Special Financing Districts" rail traclconly for the freightcrafllc~ Thus it iS where a specialtax islevied; (c) 'Exaction' offunds crtical for the company to avoid this additional orlandsfiomdevelopersandlorotherbeneficiaries. burden, whilekeeping the rail trackavailable. One 3.59 Although these arrangementsare not as exampleis the Shinetsuline whih isimportant fr 11 lisie in J as in the U.S., some the companyto tansport oil fiom Nagoyato the schem ofinternaizaion havebeen implemented Nagano prefecture. However, the JR East will and are munmi as follows: abandon a part of this line when the Hokriku a Buraen suharizngby dclaopers Tois scsemc Shnkansen startsoperation. The localgovermnt is a kind of an "Exacdon," and has several of Nagano has decided to rctain tis line and to vaation; burden sharingschemc by new establisha Public-PrivatePartnership" (described twn developerwhich has been agreedby 36 Japanese National RailwaysPrivatization Study II the Minisutyof Transport, the Ministryof PPP is discussedin the followingsection (Metro- Construaion, and the Ministryof Finance, politan Commuting Lines). burden sharingby developerswho invest in 3.62 In addition, these examplesof public in- real estate developmentsalong the railway, volvement have accorded wdl with the "vertical which is stipulated in the land development scparation" (separation of a ownership and an guideline;ad hoc agrcementbetween the operationofraiway) conceptandthe"BTO"(Build- railwaycompany and developers Transfer-Opcate) convention.The followingsim- b. Internalizationwithin a railwaycompanyr As plifiedscheme ofthe railwaydevelopmentin Japan they are engagedin the urban developmcnt illustratesthis accord; business,majorJapanese private railways by a. the JRCPC constructsa new line, usingboth themselveshave a capabilityto internalize public and private funds (Build) the externaleconomics indirecdy from devel- b. the JRCPC transfersthe constructedline to a opinga new railwayline. PPP who is investedboth by public and Goternent In volvment privatesector and receivesa financialsupport firomthe public sector (Initial Transfer) 3.60 Neverdless, since thlse arragements in c. 1) the PPP operates the new line (Opecrate) Japan are rather ad hoc and are not suffidcientto 2) the PPP leasesthe line to odter railway financesuch big projectsas the new Shinkansenor operator (SecondaryTransfer & Operate) urban line construction,thc ccntral Govrnmcnt 3.63 This arrangementhas been made Possible andlor local governmentsare exaectedto parEc- by the RailwayBusincss Law of 1986,which has pae in these infrastructuredevelopment projcts. effctvjy introduced the principle of -vertical Publicinvolvement can be classificdas follows:(a) separaion3 in the Japanese riwa business,by establishmentof a pubtic enterprise,(b) finaniQal providingthree yps of rilway busincsslinses. supports by govenmmentsto a project, and (c) As a summary,TableI11-14 illustrates the public establishment of a "Public-Private Parmership and privat involvemnt in a railwaydevelopment (PPP)."3 These examplesof public involvement i-ngJpan aim to transfcragenral sourceoffundsoftaxesand so on to the newlineconstructorandoperator.The Functmonsof the Ralway Developntm Fund dfs ofthesCtypesoftransfer are tat, (a) theral 3.64 The Shinkansen Holding Corporation, bccficiaryis not spccifiedand thus the "bencfit whichwas established solely to hold and lease these teory" is not appropriatelymaterialized, and that lincs,was dissolvedat thc cnd of September1991. (b) thercalso CXiSts an imbalanceof the levelof the Onthdebasisofthefunds raisedfrom thesalesofthc pubihcsupportamonguiwdevelopmcntprojecs Shinkac ens, the Railwy DevelopmcntFund was 3.61 Among thesC transfer mcasures, as the cstablished t implement the comprehensivesup- JNR has been Prvatizd in JaPan,railwaY develop- port policy for the railwaydevelopment. In addi- ment by a public entpr is not likelyscenario. tion,theFundisinchargeofrepaymenttotheJNR Howevera municpality involvementis stl actve, Settlet Corporation,as in the cascofthe exist- namelythe public corporationsof a municipal ing Shinkansenlines. for subway development, whih receive a hugc 3.65 The RailwayDevelopmcat Fund has two amount offinancialsupport, about seventypercent major supporing measures as follows, classified ofthe constructioncost. The secondaltecnative has principallyaccording to the sourceof funds; been broadly implemcntedby the RailwayDevel- a. The funds for the firstcategory of support- opment Fund which has been establishedto pro- ing measurcscomc from the salesof existing vide variousfinancial supports for generalprojects ofthcrailwaydevclopment.TheSlhinkansendevl- 3 Public-PrivateParnrship is named as-Sansdru' injapa- opment, which is supported both by the national anm,which liitenly means 'The ThirdSecwor.' Howmer Government and by local governmentscurrendy die Sanseku or"Public-PrivtPmtuership"'hasadifier- by up to fifty percent as a whole, is the principal ent originand dharacteiricssfiom 'The Third Sator" in the Western countrieL In geneal. "The Third Sector projectofthis alternative.The third alternative,the representsa Non-Profit Organizion in die U.S. and the PPP, has been alsoimporcant in Japan and in fact Europr,which is independentbodh fiom the firstsecrrof many PPPs have been established,especially for 'Publc"andfiomirsecondseaorofPrivate"Sane" urban lincs. Argurnents about the is norganizrionjoindyfindedby'Pubheand'Priva" andthus is profit consciousand public-interestoriented. Japanese National RailwaysPrivatization Study II 37 Shinkansensto the three HonshuJRs, which & D for the aLinear"technology, and (vii) amountedto Y9,200billion . This category investmentfor railwaysecurity improve- of supportng measuresconsisrs of, (I) a ment. grant and subsidyfor the ShinkansemDevel- 3.66 Its FY 1993 budget amounts to Y1,336 opment and (ii) an interestfree loan tO billion, of which project support budget amounts projets authorizedby the Railwy Dcvelop- to 1258billion, and almost all the rest is for the ment Fund Law.The projects for the second rcpaymentto the JNR SettlementCorporation as measureindude, (I) capacityexpansion of for the cxistingShinkmasen lines. Projecs covcred the Tokaido Shinkansen,(ii) improvenent and each budget size in FY 1993 are as follows: of the existing runk lincs, for examplecon- a. New Shinlcansen:lneconstruction (1100 nectingwith the Shnk6ansen,(ii) capacity billion) expansionof the freight railway,(iv) new b. Other trunk line development(V28 billion), line constructionin metropolitanarea, and c. Developmentof urban lins, includuigsub- (iv)capacity expansion of the urban lines,for ways(1122 billion), cxampledoublc taddng. d. Railwaytechnology developmcnt (16 bil- b. The finds for the secondcategory of sup- lion), porting measurs come in principlefiom the c. Securityimprovement investment (14 bil- generalnational budget. Measure of this lion) categoryindude, (i) a subsidy usingthe gen- erl naional budga, andi(i) an intest free 3.67 The main custoers for these supporting loan usingthe specialfunds raisedfiom the om arc japan RailwaP ConstauctionPublAc saes of NTr (Nppon Telephone and Tde- orporation (JRCPCo),TeitoRapid Transit Au- graph Co.) stocks. Projectsappropriate for thority,and railwaycomps, induding the JRs these measuresare, (i) projects of Shinkansen and die PRs. Deveopment Plan, (u) construction of local lines, (uu)resarch for developmentof Chuo TfanfrofShkaln Line and ShikokuShinlsens, (iv)improvement 3.68 TheShinkansenHoldingCorporationLzw of the existingtrunk lines, (v) constucton providedt after the leaseperiod of 30 years,all and improvementof urban railways,(vi) R the assets of the Shinknsen facilties should be

Tab-114: Publicand PfivateInvoemnt in RafwayDevelopnent investmet Subiy Char Centl G. LocalGs. Privae CentamlG. LocalGs. Privae JRs No No (Yes)* Yes Yes No Subwas Public No Yes No Yes Yes No TranitAuthority No No (Yes)* No No No NewTown Line Public No Yes No Ycs Yes Ycs PPPs No Yes Yes Yes Ycs Yes PRs No No Yes No No Yes OtherPRs No No Yes No No No Other PPPs Yes Yes Yes No No No JRCPC LocaLines Yes No No Yes No No TrunkLines No No No Yes No No PrivateLines No No No Yes Yes No Shininsm No No No Ycs Ycs No Source Japan So&ctyofC ilengineers (Dobolk Gakkai).Koux Seb Seido(Twspratwnx Drev tSySnJ). 1991. '(Yes) indicatesthatheJRs and Tnsit Authoritiesamr not ownedby dtegovemmen. but by other publicagencies # Subsidyfor subwaysofPPPs by centralgovenment and localgovernments is planningto establishin 1994. 38 Japanese National RailwaysPrivadzation Study II transferredto the three Honshu JRs. But the law Slbiikann Neworks had someshortcomings as follows: (a)Thc law did not providethe tota ransh Pcr valuc of the Shinkansens,nor the allocation 3.71 On the basis of the National Shinkansen of the total valueamong cach line or Raihwy DevelopmentAct of 1970, two seriesof among each JR. ShinkansenDevelopment Plans were drawn up in (b) The Honshu JRshad to pay the mainte- 1971and in 1973.From thoseprojects in the 1971 nance cOstfor the lines, without registering Plan,theTohokuline betweenTokyoandMorioka the depreciationfor the wholc assetsof the and the Joetsu linc between Tokyo and Niigata linc. havecbeenalreadyconstructedandarc in operauion. FinancialConstraintsfor tJeJR Central However, the 1973 Plan was temporarilyput on 3.69 The secondissueconcemningdepreciation hold in September 1984, mainly because of (a) had a major impact on the JR Central's financial mounting fiscal budget deficits,(b) onstrucion condition,because the company raises84% of itS cost increa owingto the inflation,and (c) mana- total revenuc from a single linc of the Tokaido gerialproblcms of the JNR. Then backed by the Shinkansen.During the initial three years of ics requst fiom the localconcens, the 1973 Planwas operation, the investmcnt by the company had reexamincd in January 1987. In August 1988, amounted to about 1330 billion (of which three prioriEyamong the projectsof the 1973 P1lanwas quartersare esimatd to rlate to the investment detmied joindy by the Govcrmnent and the qortheShinarnsens,w wilatheompanycouldhave then leading party (LDP), allowingfor a review registeredonly 1190billion of dcpredation, and withinfiveycars,in thelightofeconomicandsocial consequentdythe long-enr dcbt (outstanding conditions. amountofbonds and loans)had increasedby 136 DevelopmentScheme billion. While the company had been registcring 3.72 Then in January 1989, the Government one ofthe highestprofit-kvls among theJRs, ic., annouced insdecision on t developermopertr, the companyis the onlypaenger railwayJR that and fimd sources for thesc priority projects as had becn spcnding more for the investment tlan folll the internalsource of fitnd , or the amount of the depreciatonandretainedearnings,andthuswcak- a. JRCPC constructsand owns the newlines eningitsfinancialcondition.Inordertoresolvethis which theJRs leaseand operate financial deterioration, the company insisted on b. Averagcallocation of constructioncost is as purchasingthe assetsof thc Tokaido Shinkansen. follows Val-deof the ExistingVWcofD5rEnng*SibnksenShinkansen Linestl ZiCentral Gvernetoemet 35%5 3.70 Aftersomc disputesabouE the total valuc LocalGovemrnments 15% and its allocationfor each line, the four existing c. TheJRs bear the cost in the forn of Shinkansenlines, thcTokaidolIne, theSanyo line, i) the additional paymtentof Y1,083billion the Tohoku line, and the joetsu line, have been which has been alreadypaid for the pur- transferredto the JRs, JR East,JR Centmal,andtJR chase of the existingShinknsen lines Wcst, from the ShinkansenHolding Corporation owing to the re-appreciationof their in FY1991. The total valueof these four lineshas value, beendeterminedasl9,177billion,whichis Y1,083 ii)the leasepayment for the new Shinknsen billionabove the then book valuc,considering the lines after starting operation appreciationofthevalueduringthefouryears since dt The rule for the amount of leasepayment for rhcJNRwasprivarizedor the book valuewas lasdy the new Shinkansenlines has been agroedon determined.Thc total amount, Y9,177billion has in Dccember 1990 as follows:tic amount of beea divided in principle by the ratio of lease the leasepayment will be set with the limita- paymentfor theseasses among thrce HonshuJRs. tion of the net benefitamount that the op- Thc JR East owes 13,107 billion for the Tohoku erator, i.e, a JR.,would receiveby operating andJoctsulines theJRCentralowesY5,096biOion the new Shinknsen line, and will neverbe for thc Tokaido line, and the JR Wcst owes Y974 an additional burden for the operator. The billion for the Sanyoline. net benefit is defined as, JapaneseNational RailwaysPrivatization Study H 39 i) a benefitfiom operationof the new Varasow ofSbinkamcs Shinknsen line itself 3.73 Sincethen, construaon of al1the priority ii) a ncCbenefit by separatingthe todirlinc projects has been authorized anid started. In the abteast, if any, meantime,inApril 1991, the NationalShinkansen iii) a benefitfrom the side &fcron the other RailwayDevlopmenrAcrofl970wasamendedto lins connectedto the new ShinkLnsenlincr apply the basicscheme of the Shinkansendevelop-

Shinkansen Network

AsahikWhe

S

QIIM

]AoO}hebi

_~~~~~~~~~~ '. ...

AocUHarChiuo

_--IJ_C 40 Japanese National RailwaysPrivatization Study 11 ment, includingsupportingmcasures, to variations cnd of 1990, to be constructed as a "Mini of Shinkcansens,namely "Mini Shinkasen' and Shinkansen' which was cstimated to cost about 'Super Express."Standard Shinkansen or "Full V60billion . Then the plan was eviewed,taking Shinkansen"represents a linc with an infrastruc- into accountof hostingthe 1998 Wintcr Olympic cureof the opemacingShinkansen standard (maxi- Gamc in Nagano,and has ben decidcdto be built mumspecdof260kmperhour)."MiniShinka6nsen asa"FullShinkansen'whichiscstimatcdto ostsix representsa line with a gauge of the operating timesthe previousestimation. Thus the construc- Shinkansen sizc, added adjacent to the existing tioncosthas beenincreased,whereas thcamountof crack, but with a train of small size (maximum theleasepaymentforthenewShinkansenlinesby speed of 130 km per hour). "Mini Shinkansen thcejRshasbeensetunderaruleasabove,whichhas scheme has been developedto minimizethe con- little rlevance to the constructioncost. stuction costby utiizing an existinginfrastrucure e i TheSkinkan Development of tnk limes. Super Express' representsa new pI= *f197.3 line with a narrow gauge but with all the other infrastrucure of the "Full Shinikanen" standard 3.76 The Government decided in February (maximum spced of 160 to 200 km per hour), 1994 to postponereviewing the developmentplan consideringa future project of a gaugeexpansion of the remainingprojecas 4 in the ShinkansenDe- for the "Full Shinkansen"standard. "Super Ex- velopmentPlan of 1973,which havebeen tempo- press' standardis used for developingalinewhich rarily put on hok; since 1982, until the currendy has a direct servicewith ordinary trunk lines at undertaken Shinkasen projccts are 50% coDm- present, that have a chance to be convertedinto a pleted (in about threc to fiveycars time). Before "FullShinkansen" standard. then the policycommittee of secretary-genralsof undere ofSkinkae Consmuaion therulingpartes (ofthccoalition governmcnt) had reexamined the reaining projects and recom- 3.74 Outline of the whole projectsunder con- mendedat the cnd ofJanuaiy 1994 to construct all struction, namelythe Tohoku line, the Hokuriku the projects in the 1973 Plan by the "Full linc,and the Kyushuline, is shownin Tablc 111-15 Shinknsen" standard.In addition,the committcc below. has recommended that the "Mini Shinkansen" standard of the Tohoku Shinkansencurentdy un- 4 Teem nini projectsin the1973 planare the Hokkaido 3.75 Estimatedconstruction cost ofithcaurent lincbetceen Aomori and Sapporo(about 300 kin).the Shinkansen projects under construction, amounts Kyushuline bodt beten Fulkuokand Nagasaki(about to 1,795 bilion (seeTabldII- 15). Among dse 120km) and betwen FukuokaandYamtshim (abour 130 tooj bils lion(e -ablell- Amrou og the k'n), and the Hokuriku line boEhbetween Osaka and projects, the Karuizawa- Nagano route ofthemcKanumn (about240 km) and betweenNagano and Holuriku Shinkansenbad becn planmed,until the Nisbi-iugi (about170 km).

Tabk MI-15:Shinkansens under Consuction Mode*of Line Route Shinkansen Opeator EstimatedCost (I Billion) Tohoku Morioka- Nunakunai Mini 38 Numaklunai- Hachinohe Full 237 Hachinohe- Aomori Mini 112 Total JR East 387 Hokuriku Takasaki- Karuizawa Full 201 Karuizava- N2pr. Full 387 Uozu- Ioigawa Super 188 Nishi-isurugi- Kanazawa Super 174 Total jREast and West 951 Kyushu Yatsushiro- Nishi-kagoshima Super JR Kyushu 457 Total 1,795 * finir iepresentsa 'Mini 5hinn. -Full- reprsents a -uF hinkansen.-and Super epresnts a Super Eess. *JR Ea is in chargeof the area ea ofJoecsu.and JR West is in charge of th ar wes of Joeu. Japanese National RailwaysPrivatization Study If 41 derconstruaion shouldbe convertedinto the"Full certainamountofrevenuefortheexpansion invest- Shinkians standart. ment with a tax exemption. 3.77 Nevertheless,thecommitteehadspecified Nw LineDevelopment and PPPs neither consrction period, nor priority among the projects, nor the fund sourcr. Construction 3.79 APPP has longbeen afamiliarscheme for costhasbeen estimated bythecommiLte as V5,000 urban railwaydevelopment inJapan, especiallyfor billion, whichshould result in an additional huge the development of monorails and commuting burden for the nationalbudget, as the committee lines for new towns, whereasa public corporation has recommendedthat the burden sharingby the has been popular both for developmentand for Government and local govcrnments should be operacionof subwaysin the metropolitan areas. augmentedto alleviatethe financialburden for the Owingto a remarkableincrease in the construction JRs,setas fifiypercent for the currendyundermaken cOstsin recent years, a privatcesector involvement projects.Negotiation for the national budgetas for has been encouragedfrom a fund raisingpoint of these projects thereforchas been rough under the view.Besides, owing to the sprcadingout of metro- currnct fiscalbudget condition. politan rcsidents, metropolitan railway networks havc become much more complicated and it is requiredto negotiate among many concerns to Pnv RawysPe and CapacayExpanion fiuher dcvclop the network Thus it has becn 3nareaofjapan, the PRs decided to accelerate the cstblishing of PPPs, h3v a dominantshareofmabout two irdsin tcrms whichhave bettr accessto publicsupport than the hf rout inngth,agair oahbJRs. Ncvcrtheirs sinsc PRsor theJRs. The followingare severalexamples the construction cost of a new line has increased fromcurentor plannd projcts whichareunder- dramaticaly, the PRsare barelyable to construa a 3kcnbT PPPso new full line by themselves.However, congstion maes the pubic support called P Line Mthod in trains has becomesevere in the metropolitan kstpbc poadPne td area,especialy in the morningand evening, reach- 5 'P LineMethod whichacoordswth aBTO sdheme bydie ing as high as two hundredpercent of the designed !RCPC,has beenapplied for a nEwline constructionin capacity.Thus the lawto promote railwaycapacity metropolian2reas.Thissupportschemeindudes (a) allow cxpansioninvestmnt expansion in metropolitanmceropolitan areas was ~~~JRCPCatRn&ireeoperatorroin 25 yearsafte redeemtheonsructioncostto the operationsum, and te(b) enacted in 1986. The laswallows those PRs, that subsidizeth cost for inettr paymentwhich exceeds fve have a plan to expandrailway capacity to set aside percent.equally by the nation and the localgovernammt

Table15-16: New Lne Devdopmnt Constuction Constmction Cost Length Operating Name of Line Peiod (YBillion) (kn) Entity LocalGovements TokyoSubway#12 1989- 1996 683 28.8 Tokyo Tokyo Rinkai Fukutoshinsen 1991 - 1995 128 4.9 PPP Tokyo Joban Shinsen 1992 - 1999 800 58.3 PPP Tokyo, 3 others SaikamaSubway #7 1994 - 1999 259 14.5 PPP Saitama,3 cities MM21 Line 1992 - 1998 195 43 PPP Yokohama city Chiba Express Completed in 2000 93 10.9 PPP Chiba, Matsudo city Toyo Express 1984 - 1994 220 16.2 PPP Chiba, 2 cities Kamiiida Line 1995 - 2000 70 3.1 Nagoya Aichi, Nagoya City Tozai Linc 1989 - 1995 61 3.5 Kyoto Kyoo city Kaafiuc Line 1988 - 1996 330 12.3 JPWest Osak2, Hyogo, 2 cities Osaka Sotomnjo Line 1994 -2000 135 20.4 JR West Osaka, others Sourc Chiho Koci igyo no Shin Tenkai Nado ni KanrsuuKenkyuu Kai (Study Commince for the Local PublicEntise Devdopment),Tosh TewnAS nwSei&i no Arikaa i Twire (Su. on UrbaxRaiuay Deveipmnen)December 1993 42 Japanese National RailwaysPrivatization Study II insufficientto make projectsfeasible. Average con- was establishedin March 1991, by the concerned structioncost ofa new line (seeTable 1II-16)in the four localgovernments and twelvemunicipalities. centerofmetropolitansexceedsV20 billionperkm. Capitalfromtheconcernedpublicseccoramounted Even in the suburban area, it reaches about Y10 to '130 billion, whereas that from the private billion per Ikn. The Chart III-5 illustrates the scaor amounted to '30 billion. The PPP was feasibilityof the typical projects currendy under- licensedinJanuary 1992 under the firstcategory of taken in metropolitanareas; the Joban Shinsenin railwaybus.ness to awhorize a companyto operatc Tokyo and the Katafiku Line in Osaka. This a railwaybusiness wiLh its own track analysis indicates that further financial support 3.85 The current estimate of the railwaycon- from the nation or the local governmnentsis re- struction costamounts to 1800 billion.The PPP's quired to construct a new line in a metropolitan paid-in capital both from the public and private area, given the current tariff system. sectors amounts to 160 billion, and the PPP is 3.81 Inaddition,thcsecwoprojecctsoftheJoban currently raising funds for this project through Shinsenand the KatafukuLine showa good con- interestfree loansboth from the RailwayDevelop- trast in tcrms of theJR's attitude to and its involve- ment Fund and from the localgovernments. Nev- ment in the project;while the JR Easthas declined ertheless,since the cashflow requiredis inaeasing, to own the track or to operatethe Joban Shinsen, the Ministry of Transport considers to provide theJR West hasbeen eagcrto operatethe Katafufl loans from the FILPs(Fiscal Investment and Loan Linie.The rest of this sction discussesthis issueof Programs)from FY 1994, whichin turn isexpected the JRs' involvement. to atta investorsfrom the privatesetor. a. JobanSinsen b. atuafiu Line 3.82 Mhisprojecc was fist recommendedin a 3.86 The project of a 12.3 km new line was reportbythe"CouncilforTransportPolicynofthe authorized by the Ministry of Transport in April Ministry of Transport in July 1985. This new 1981. This line will connect direaly the east and urban commuting line with a 58.3 ianlength will west part of the Osak mtropolitan area, via the combine the Tsukuba city, one of the major re- central businessdistrict of the city through under- search cenrers in Japan, with the center of the ground. In other words, this project is the key for Tokyo metpolitan (Akihabara),by 45 minutes. the urban railway network dcvelopment in the The major purpose of this project,however, is to Osaka metropolitan area (see Chart IV-2 of the promote residential developmentalong the line, raihay networkin Osaka area). JR West has been constructing about 170 thousand houses. kEis alvays enthusiasticabout this project, although expcaed to start operationin FY 2000. this is one of the most expensivelines currendy 3.83 The report also recommendedan esab- undertken in Japan. In addition,new town devel- lishmentofa PPPfor this hugeproject. Then a bill opment has been encouraged around the Osaka concerningpromotion ofJoban Shinsenhas been area, specificallyin Sanda city whih is locatedin passed in June 1989. Neverthdess it took a long the cxtcnsionof this line, and thus made this timc to formulate a PPP, because of a growin6, project further prospective. cautiousnessofJR Eastwhich was assumedto be a 3.87 Thcotherkeyissueofthisprojca istheuse major participant.JR East evenmallydedined to of a PPP. A PPP called 'Kansai KosokuTesudo own the tcrackand even to operate the new line, (KansaiRapid Railway)"has been establishedin basedon thefindingsofa feasibilitystudy,cxumin- May 1988 to construa this newlineand to leasethe ing the profitabilityofthe linewith a then assump- track to theJR WesLThe principalinvestors arc the tion of the construction cost of Y700 billion (in Osaka local goverunent, the Osaka municipality, 1987price).TheargumentsbythcJREastwasthat and the JR West who own 22.9 percent of the the company cannot transfer the hugc develop- stodks respectively.As a whole, thc public sector mcnt cost to thc generalcustomers of other lines and the privatesectorequallysharethe stocs The andthat thosewhocangaintheextcrnaleconomies major private investors arc dic Kansai Electric from the railwaydevelopment should bear he cos PowerCo., the Osaka Cas Co., and other compa- and risk of the project. nies based in the area 3.84 Then at last,a PPP called ShutokenShin 3.88 The principalincentive for using a PPP is Toshi Tctsudo (MetropolitanIntcrcity Railway)" to realizebeter accessto the intercstfree loan from Japanese National RailwaysPrivatization Study II 43

Chart111-5: Feasibility of New Lne ConstructionIn Meopolitn Areain Japn: IllustIe Analysis

7501- u 2.5 0 3600. >' 2 -

= " - , -- ^' ~~~~~~~~~~~~~~Capital Cost**

30D. -L._

151 0.5 Jloban Shinsen'

5 10 15 1 b ' -' Construcion Cst (billionyea/km)

Sourc Producedbased on nJa Ski sn xi TsakeKaagarm(EJ=widg theJohaaSbiMeaM)bYMaSarkeMatrsuda. Unyu w Klzai (Transporation & Economy).Ocober 1989.

Notes Operating Cost is assumedas VW.41billion per Iam.based on the averge figuc ofthe majorPRs in the Tokyoarca in FY 1991. CapimalCost is calclated assutming,inprnciplekP LineMchodw that (a)30 percentoftheconstructon cDstis coveed by paid- in capitalwhichrequires 10 percentROI (reurn on investment).(b) allthe other financeissubsidized ifthe interestrareeeds 4 percent.and (c) the assetis deprciated in dtirty yes TrafficDsity for theJoban Shinsenassumes 189 thousandpassenger-krn per rouwekm per day. basedon dth for the other suburbanJRline in the Tokyoarea. Traffic Densityfr the Kaafizu Lineassumes 300 thousandpasger-km per routc km perday. basedonthatf(orthedensstsubwaylinint heOsaaandTokyo metpolitanateas 353 thousandfortheTozailinc in Tokyo.343 thousand for the Chiyoda line in Tokyo.and 334 thousandfor the Midosujiline in Osakb Revenueassumes V9.2 per passenger-km.based on the avea figureofthe majorPRs in the TokyoMetropolitan area in FY 1991.

the govemment'sspecial fiund of the NTT stock 1ntrcyLins sale an the JR itsef does. Equallyimportant, a PPP cm arrange smooth negotiationamong van- 3.90 As the railwaynetwork in Japan has been ous concerns to promote the construction, since areacdywellestablished, there are onlyfew projects thoseconcerns are aually invohledin the projects to constructnew intercityrailway lines, except for as investors.In addidon, JR West has not had an the Shinkansens.Nevertheless there stillexistlarge experienceto construct a subway, fir which the needsfordhespeedingupaadcapacityecpansionof Osaka municipalityhas a good pienc the existinglines. The followingprojects show a 3.89 The construction had been temporarily basicschuemeforaiwaydevelopmentandmanage- haltedbyan accident,and its inaugurationhasbeen menr fimd raising,subsidy from the Government postponed from FY 1995 to FY 1997. And the (the RailwayDcvelopment Fund), establishment construction cost has been augmted from the of a PPP (ocal goverment involvement), and original Y288.5 billion to 330 billion. By the opcation arangement. originalinvestrent plan, 61.4 percent of the total £t YalUa ShY costis coveredby theJRCPC,and only7.4 percent is coveredby the loan from privae bank The =st 3.91 Althoughit is ernedas a Shinknsen, the is covered by the entity's own capital and other Yamagata Shinkansen is not a full standard public financal supports Shinkansen, nor is it included in the national 44 JapaneseNational RailwaysPrivatization Study II ShinkansenDevelopment Plan. This projeacaims c. Se&irhoLine to achieve a speed up of the intercity transport 3.96 Thisproject wil upgadc one of the major betweenYamagata and Tokyo,through (i)upgrad- 3.96sthejRokkadontrofuchew ing the existingtrunk line betweenFukushima and trunklcnesoftheJRHoakaido and introducea new Yamagata(90 kmn)and (ii) connecting the trunk expresstralinwithanamaximwspeedofl301cmper line directlytO the Tohoku Shinkansenat t hour. Thus it will reducethe tir.e requiredfrom Fukushimastation Sapporoto Kushiro by 40 minutesto 3 hours and 3.92uThemptaost 40 minutes. The route length upgradedby this 3.92 'Me project cost arnountedtO V51.3 bil- Ron, induding Y24.0 billion investmentfor the pro)ect reaches304 km. ro,ings,odis. In ordcr to construcEa tmck and m 3.97 The construction cost amounts to Y10.7 rollingstocks. In order to constructa track and tO ilon e...d.n roln st..i.stn bor.b purchasemoling stodcks, a PPP called 'Yamagama JR bglilon,excluding moUing StOCK investmnent born by ChokkoToklknu Hoyu (YamagataJR Direca Ex- the JR Hokkaido. A PPP called "Dot Kosoku prossHolding Co.)uwas mstablishedin May 1988 Tetsudo Kaihatsu(East HoklcaidoExpress Railway and was capislized equally sythe local gove1- Development)"was established inJanuaty 1994 to mendsand theJR Easlgroup. theRailway Devel- construct and lease the track to the JR Holdaido. opment Fund provideda subsidyto the PPP with 'MeJR Hokkaidoand a group ofthelocal govern- twenty percent of the constructioncost approved nnts each providesV2 billion for the capitaLAnd by the RailwayDevelopmcnt Fund. the RailwayDevelopment Fund provides12 bil- 3.93 Leasingthe track and rolling stocksfrom lion as a subsidy. the PPP, theJR Eastbegan to operateehe upgraded Issues of Freight Transportation line in July 1992, connecing with the Tohoku A m entofi'doISI* _hinknsen.The time requiredfiom Yamaaa tO Tokyo has been reducedby 42 minutesto 2 hours 3.98 It was a generalperception at the time of and 27 minutes Thus the railway has become theprivatizantionthat the railwayfrwightservedon compettiveagainst the airjourney,wiich takes 55 anarginal rolc in theJapanesefrcight traffic market minutes for the sametrip. The number of air and that the profitability of railway freight was passngrsbetweenYamagandTokyodecased diffa tO justif. Howe such peoenon byamundonequarterinthefirstninemonthssince cmergedjust aficr the privatizationas the rapid the YamagataShinkansen staed operation. growth of the rilway freight demand, the traffic HokuetjuN,rtb Line congestion in the road, and the labor shortageof trudc drnvers,cncouraged the argument of modal 3.94 Theconstmucionofdtisnewlocallinewas shift Besides,railwayr fright is thought to havean resumedin 1985to makeashortcutintheroutefiom advantagein terms of cnergyefficiency' of trans- Tokyo tO SUChmajor itiesin the north coastof the portation and of environmentalcxtcal cost, al- Honshu (mainland)as Toyamaand Kinazzwa.The though the advantge has not yet been wholly route lengthwill be shortenedby about 55 km, cxaminedin Japan as has been donc in the argu- dtough this new bypassrourc of 59 km, and the ment of "infrastucturc frie in the Nordic coun- time required will be reducedby 20 minute 3.95 APPPcalled"HokuctsuKyuko(Holkuesu 6 Advange incenegeElciencvyofraihafighaainst Co )"owasestablished inAugust 1984. The de roadfieight has bee broadlyadvocatd in Jyan. localgovecrnrcnts ancd prvat companieshold 83 Neverthels someprofound smdies on this issue do noc nelissarysupport thisgenerI percption. when the rail- percent and 17 percent of the stocik respectively wayfieight is compuredto the roadfreight of thelong and the amount of the paid-incapital is expcd to distance,which is the ral competitorof the raily. For be increased to Y4 billion beforc the operation etample.Hidala condudes tat on ihe basisof line haul starts.In addition,the R ailwayDevelopmentFund enerW.which indudes both mnningaerg and indirecr enezgy,the propottion of ecrgyperwon-kniconsumed by providesa subsidyandtheJR East providesVI5.8 t railwam dr by the tuck is 100 w 70-80 and dth billionas a burden chargeto the PPP. The cost dweis not so much differnce of enegyefficienq C"En- amounts to 24.0 billion for the new route con- ergyEconomicsdv&oL19 No. 16byrhelnstittofEneraj suuoon,andupgcdig costofthesciscingluu*Of Economii,Japan). Contary. the road feight ofthe short theJRE:astamounts to 179.4billion .AndthePPP thean i heinno rcigtyraffic,isthernjor probkm of ,,,~~~~~~~~~~~~~~~~d temfieanqin ofdh fixght r&

Tabk1117: Stabs of Seprauionsof theNational Ral7ways byCowlly VericalSepation Line-of-BusinessSeparation Regionl Sepron JapanURs) (No)** Yes Yes U.S.(Conrail) (No)** Yes No Canada (CN) (No)** Yes No Sweden(SJ) Ycs No No Norvay (NSB) Yes(by financialseparation) No No Finland (VR) Yes(by financialscpartion) No No Swiss(SBB) Yes(by financialscparation) No No Austria (OBB) Yes(by financial scpaation) No No Gemany (DBAG) (Yes)*** (Yes)** No Son= Tenod* Je&rw'ii ThduS furox'iay)R byMamichiHori.UnywKeizi(Transpormtion&Economy). April M9. Others Notes: Line-ofusinessSepamtion ia separationby a railwayand a friht ilay business (No)indicates itac the ra trak arenot evaial seprfrom a h raiy. butare sepaated from some rilwy. DB (DeuscheBundesbahn) and DR(Deuc ieicsan)in Germany had bem separed. butwasunified in jantury 1994. Ra =rckholding, the freightbusines and pasenger businessare se ndnside DBAG (organizational separaon) and are expectedto be compketdyseparaed and managedby individualcompanies in 1997. Japanese National RailwaysPrivatization Study 11 47 the long term one. In addition. the avoidablecost competition,we comparethe operaing costper ton- still has such a disputablcissue as exdusion of the lcn (seeTable III-18). The costof railwayfreight is direct labor cost for maintenance cost (only the almosthalf that of road fieighc.It is estimatedthat materialcost and subcontractcost are induded). curdy railwayfreight bas advantageon cost be- .. mpeuisionT. .ith .krandl,adicatiueEcenom,tweenTokyo and Osaka, and thc rason why freight eo~ontnrlu haveeceeded railwayfreight is the inconve- nienceand the weakmakeing ofJR FreighL 3.109 Theargumientofthemodalshiftandthc 3.113 Second, we comparethe operating cost aaual environmentsurrounding JR Freightinevi- on equal footing, which cxcludecross subsidiza- tably arouses the discussion of the viability of tion, andwhich do not includeconstruction costs. freight train transportaion in Japan in the long In the caseof roadfreight traffic,dhe costs of road ntn. The centralissue here is the competitiveedge constructionand maintenanceare consideredto be of train in comparisonwith other traffic modes, subsidizedtosomeextentbypassengercars.Therc- especiallywith tmcks.The issue iscomplicated by fore, 1) we exdude tax (induding oil tax) and theallegedcrosssubsidizationeistingboth intrain highwayfee expenses,by which tiuc share road and truck transporation. The issueis fiurther com- constructionand maintenancecost, fiom the oper- plicaed if extnalities such as pollutionand con- atingcost. 2) weadd estimatedniintenance costof geiion are ake into account. road,inwhichtheburdensharinofpassengcrcars 3.110 Inthelasccoupleofgenerations,theedge is eliminated (assumingonly trucks pay it), be- of fret trudcshas =dedthat offreightrilway, tweenTokyo and Osaka.In caseofrailay freight though the ilwayis considered to haveadvantages traffic,we assnumcthat the real avoidablecost (linc overthe trucforlongdistance frcighThc ason is leasepayment ofJR Freightto passengerJRs) is 4 notconsideredtobethacthecosofJRFihrishi times as much as the line las paymentwhih JR but ratherthatthemadngofJRFrcighcisweak. as FrdghcurrendypaystopassengerJRs.Asaresulr,the is discussedin the elie section.To compar these oper cost on equal footig of railwayfight iS two ficght modes,the costsof dte road feit and lowerthan that of oad fieight.Although the esd- railwayfreight arc calculatedin tbis parm matedoperadngcostonequalfoodngisverysensitive, 3.111 The costcomparisonof roadfrightand it maybesafetO saytatraiwyfit can compete ilayfrightisdifficultforthcfollowingreasons, roadfreight on costwithout cross subsidization 7 on the one hand passenger railwayshave given advantages to JR Freight based on the avoidable Ralion ofthe Goment cost principle. In addition, it is said that the real Priwazon and Rgulaion avoidablecost isgreater than the line leaseepee 3.114 Alhough regulationsare imporant to whichJR Fr t acually paysto passengerJRs. On sre sound managementofand safeoperation by the odhr hand.it issaid dat fight trucls do not pay 7 EstimatedCsCompasonindungC onscruaionCosb the appropFriateosts for theiroad constuction and Whenbualding and extnding freighttanspor network disrAcion. For example,the oil ax is specificalyone musecompare cos includingopeatng cosrand spent to the road consuction and maintr.iance consuctioncosr-Bothrodanddilwayline ostruccion However,tax for light oiL whih manytrucks use as cost which exdudeburden sharingkr passengercars and fiel inJapan, is cheaper than tbat for g ine,whrilways mustbe considered.Consruction costperton-ln of road fieightis. 114 that o fralways. sincerod usageis 5 manypassenger cus Moreover,trucks are suid to t imes= dun that of raiway tracLTotl cost of road distractroads in the order of tens of thousandomes f&ieXis 1/2-113dut of yilv fieight.so nihay freight more than passengercars do. annorcompeewith madfreighton cost if freighrrailway 3.112 Here we focus on one route, between alon raherthan both freightand passengr railway is Tokyo and Osaka (580 km) - becausebetwn thereson for constructionof newlines. See Table 111-19 Tokyo and,Osaka.1), freight de, and is bl . 60followir.This estiaut isbasd on specificconditions of Tokyo and Okah, 1) ficot demand IS biggcstin +figtransportation injapan. Firstthe trackusageofJR Japan. and 2) alhough it issaid that a mmuingpoint Freighis arcsti d bypasngeJRs buse of high train of trucksand railwayfreit is 400-600 kin, the fiequencyonsheTokaidoline.Iftheae isno estricton.jR shareof railway(5%) is much lowerthan that of Frightcaninihassenumberoffreiihttmins.Second, wiclk (51%) - in order to estimate the cost consuction costofroad nd railwayisexrmely expen- sivein JapLn dueto the hihland priceandlaborexpenss cDmpeauonofmad fit& and ihway eidghtwz& Therere the compettionin other countries may be outsubsidy.Pirs,inordertocomparethecurrcosr diffeent fromthe one in Japan. 48 Japanese National RailwaysPrivatization Stcy 11

Table111I18: The estimatd costcomparison of theroad freight (ToMel highway and Meishin highwy) andraflway freight (Tokaido line) between Tokyo and Osab (IVperton-km) Road Railway OpeatingCost (A)8 14-16 8 Tax and Highway Expenscs/ Line LeasePayment (B) 4-3 0.8 EstimatedMaintenance Cost (C) 1.1 3.2 Opeating Cost on Equal Fooung

(A)-(B)+(C)8* 11-14 10.4 ' Operating Cost * Operating cost for road, for the long-distnce uuck with 10-20cloading capacity. is basedon dth data in 1983. We aumed tharcost incased from 1983 to 1991by 139. (Sources"Tokkdcu no Ogaah ni KansuruChosa Kenkyu HokokuzSho (Smdy ReportonLarger-sizeTrudcs)." 1983. InstituteofTransportaion Econo mics, 'JidoshaUnsoligyo KeieiShihyo (Financl Data of Auto Expressi gent). by Minismyof Tansport) The operting cost esduds cosmsfor pickup and deliveryof cro, and renninal. * Opating cost for ihwayis the awrageopeatingcost ofJR FreighAverage opang costis influencedbythe fright taffic distance. However.averge freighttraffic distance ofjR Freightis 470 km and dte distancebaween Tokyoand Osaka is 580 km. Therforethe avcrageoperatingcou betwenTokyo and Osakais consideredto bealmostthe sameas the aveage operxung cost ofovera1JR Fright. Operting Cost on EqualFooWtin Operating cost whichexduded co= subsiditaion * Road.Cross-subsidition for the tudcs is consideredto bemainlyroad mntenanc costad road consuction cost ofwhidc pasnger carsshre largepar. We subtractthe tax (includingoil tox) and highy fees 4 - 3). by whidcroad are maintained and construted. fromoperatingcst Then we addtheesdmared maintnance cost(I1.I) ofhe road betweenTokyoand Oska. asing only trus pay it.

Tax and Highway Expenses V per ton-km 4 - 3 Mainienauce Cost *1 (D) Y million per km 80 Road UsaW **2 (E) million ton per km 73.5 Estimated Maintnnce Cost (D)/(E)V per ton-km 1.1

.1 Maineena (x.t: The naintnance cost for Tomei hihway is not publicized. Her it is assumedto be as mudc as the maintenanc costot lnhway in metropolitanaras such asTokyo and Osakabecause the nufic dasit of Tomei hiwy is high though the Tomei i,thway is opemri by Japan HighwayPublic Corporation. Japan HighwayPublicCorp. 147 miltion/ km MetropolitanExpressway Public Corp. V176million I km Hanshin ExpresswayPublic Corp. V165million I km MaintenancDcstis smallertun thecoabovevwhich indude improvementexpenseandrestoration apense (SourceJapan HighwayPublic CorporationAnnual Report) `2 RoadUsg RoadUsage. at 73.5 millionton per km perycar,is calculatd by dte tI trudc operatinglength on the Tomei higway (smallfizight trucks 3,371 thousandkn per day. nxdium and largefeight trucks 13.559thousand km per day) and theaverageloding ofeach tuck (smallfeigh trucks0.6 ton. mediumand lage freighttrucks 5.0 ton). (Source `Taskso$ he 12th National Urban Expresswas for 1990 Fiscal Yeae by Fukluoka-l(iakyushtExpnssway Public Corporation fiom 'Exprssways and Auromobiles"5/92 and 6192, ExpreswayResarch Foundationof Japan) Railway Line L se Payment "3 (F) V millionper km 103 TRck Usage 4 (G) millionton per km 13.7 Line LeasePayment (F)I(G) Vper ton-km 0.8 Esdmated RealAvoidable Cost "5 Yper ton-lkm 3.2 3Line LeasePayment 320 billion(the acual line leasepayment ofJR Freight)x 0.3 (assumingthat tonnageofTocaido line is 30% of the tal tonnageofJR Freight)/ 580 km (route ength ofthe Tokaidoline) - V110.3million per km per year. "4Railway Trdc Usge. at 13.7 millionton per km per year betweenTokyo and Osak. is calculatedasming tha theonnae on the Tokaido line is 30% of the total tonnage of jl Freih 26.3 billionton-kin x 0.3 / 580 km - 13.7 million ton per km. "5We assume that real avoidablecost of JR Freightis 4 times as much as the line lase payment. Japanese National RailwaysPrivatization Study 11 49 railways,considering that the JRs are going to be for the JRs. The major points of the rdevant fitUllyprivatized or sold to the public, it is also articesin the BusinessLaware, (a) theitriffshould importantto minimizethe regulationsor at leastto compensatethe adequatecost under the efficient discloseprocedures of these regulations.The pri- management,and (b) the tariff should coverthe mary legislature for the railwaybusiness is the appropriateprofit margins.The first problem on "RailwayBusiness Law" (the BusinessLaw hereaf- this tariff issue is that these conceptsof adequate ter) which stipulatesrailway companies' develop- cost,efficient management, and appropriateprofit mentplan,operation,generalmanagementofralvays, marginsare not preciselydefined and are vulner- technologicalstandards, and so on. Neverthelessthe able to the occasionalgovernment control, or PRs are also requiredto followthis law, and thus politicalconsiderations. Furthermore, since op- these regulationsare not specificfor the JRs. erational environmentis so different among the 3.115 The LawConcerning Passenger Railway JRs,and betweentheJRs and the PRs,it isdifficult Companiesand the Japan Freight RailwayCom- to presume a benchmark for these management pany stipulatesregulations specific for the JRs. indicators. Clauses of the law require an approval by the 3.117 Theotherissueonrhetariffsystemisthe MinistryofTmnsport,concerning external financ- timing of the tariff revision.The experienceof a ing, appointment of CEO and auditors, annual PRs' tariff revisionshows that a tariff increaseis businessplan, ransfer and mortgging of impor- permitted (ice.,has to be acceptedby the Minisuy tant property,revision ofthe artides in the incorpo- of Transportation (MOT)), only if the railway ration, diposition of profit or loss,and mergerand companiesmake loss or are expectedto make loss dissolveof the company. All the clauses huted fromtheirrailwaybusinessNcvcrheless,itshould above,however, are also important items for pri- be noted that the profitablydiversified business of vate companieswho are rcsponsiblkfor the share- PRsmakcstisurvisionrulepossiblecassuringtheir holders, but not for the Governmnct. Thus the profit bases.On the contray, dcivesifiedbusiness bottom linefor the regulationargunent iswhether of the JRs is in its infancyand can not contribute the Governmentconsiders the importanceof the sufficiendyto their tota profit.As for the electric- sound managementof the JRs as much as the icy tariff in Japan, operatingloss of public utility generalpublic docs for privaterailway companies. companiesis not a prerequisitefor the tariffrvi- Among theseitems, the most straightforwardof sion. regulationsaffecting the JRs' managementare (a) 3.118 Tariff revisionrequires the approval of ytriffscm, and (b) investment. MOT. However,currcndy Japanese goWrnmcnt TariffSjvm is promoting deregulationof domestic industry. Asparcofthiscffort, the Governmcntis dcreasing 3.116 As forthe tariffsystem,the Businesslaw the number of regulationsthat MOT controls, providesa commonprinciple both for the PRsand includingrailway, road, airwayand ship, by 20% Table11-19: Cost Comparison induding Constuction Cost Road RaiW Construaion Cost (H) (Ymillion per lum) 962 770 Road Usage{I) (millionton per kIm) 73.5 13.7 Est. ConstructionCost on Equal Footing (J)= (H)/I() (t per ton-km) 13.1 56.2 Operatng Coston Equal Footing (K) (Vper ton-kIn) 11-14 10.4 Total Cost (J)+(K)CYper ton-kin) 24-27 66.6 NotsTeaveage costforroad(962millionpcrkrnperyear)andforriway77o millionperkmperyear) arecalkulared assuming dat: a)consmtuion cos entils 5%inso expense2mnualiy; b)asse aredepredated in 30 years;andc) unit construction coDs are thefollowing: VI 5 bilion peri kforroad (thesameunitcostastheTokaidohihway and Vl2bgionperin forraway(ieavaage unircost of raiW constrctionis sihdty less than the unit cost of roadconstmction.) Sourc JapanHighway Public Corporation Annual ReporL 50 Japame National RailwaysPrivatization Study II within 3 ycarsfrom 1993. Regardingraihway tariff RclationsArrangemcnt Law) was enacted to be system,in 1994 MOT is expectingto amend the applied mopublic corporations. Under this lanw, BusinessLow that railwaycompanies can discount JNR employeeswere given the right of collective tariff of faciliy services,such as firstclass seats and barg.ining, but were nor allowed to strike. Union sleepingberth,withoutapprovalofMOT.JRs and membershipwas on an open-shop basisand work- PRs can increase or decrcase tariffs on facilitY erscouldchoosenot tojoin, but most didJoin.The sevices at their own discretion. NationalRailwayWorkers' Union, orKoktro,was InvestmentDecison the biggestand most powerfil, and representedall typesof workers.Politically, it had dosc ties with 3.119 Theissue of investmentreatues mainlyto the Japan SocialistParty and frequentlyled illegal the constructionof new lincs, especiallythose for actions for the right to strikc in the 1970s, which bullet trains or urban transpot In both cases,a turnedthepublicapinst]NR, Three other unions hugeamount of funds is required.Since a railway split from Kokuro and duly registerd in order to company inhcrcntly has two different negotiate with the management. Thc National characteristics,thatof a profit oriented business Motive PowerUnion. or Doro, representedsolely cntityand that of a public servicesupplier, invest- motive drivers,but thercwere also many motivc ment decisionsfor railwaycompanies must cake drivce who remainedwithin Kokuro.The Japan acount of the public interest,and are likelyto be Rilway Workers Union. or Temsuro,was politi- vulneablc to politicalpressure. callymilder and supportedthe Japan Dcmocratic 3.120 The history of the upgrding of the SocialistParty. The All National Railways'Pcmna- Hokuriku Shinkansenline showsthat the operator nent`WyandConstructionLaborUnion,orZ7ensirf, (in this case JR East) was fidly involved in the polirt supporedanodtrcenrimparty,Kreito. decisionmaking processfor the newline constrc- 3.123 JNR manemen negotited with cach tion, which required a huge amount of finds. ofthescunionsissucssuchaspayandcbangcsin Otherwise profitability or effective opeational train schedule, though pay rises usually ended in managementwould be asured. Another example the same incrase as was recommendcdcach year of operatorinvolvement in the investmentdecision fur public servantsby Nationat PersonnelAuthor- nakdngprocess is the Joban Shinsen,a 58.3 km ity.Thesalasystandardwasinprincipleuniversin long new urban commuting line in the Tokyo all CaE with minor differentiation by type of metropoirn area. Afer examiningthe feasiility crafts. Rct- zment age was decidedby agreement of the new line, JR East deided against being an between the managementand each union and was operatoror an invcstor.Thus the projecthas bcen basicllythesamcasthatforpublicservants. Politi- initated by a PPP. waly,the leadersof Kokuro and Doro were influ- 3.121 Since the JNR was privatzed, the JRs encedbyMarxism,butinmanycascsanewworker havehadto be responsiblefor theirproficabilityand tended to choosethe unionwhich was dominant in financilmanagment. A thesam time, howver. the divsion or district he was first assigned to. railays representan important publicgoods in the Kokuroopposed to the break-upand the privatiza- national coonomyand socicq. Suchabig projectas tion ofJNR until the last minute and subsequendy the Joban Shinsenis a good exampleofimuproving collapsed.The other duteeunions chose to support nationalwfire. Thus there alwayseists a conflia the break-up and privatization. of interestbetween dtat of commerial entity and 3.124 In the final days of JNR, three unions that of publicservicc Carrier. But as is indicated in (exceptKokuro) inaugurated theJapan Confedea- the above two cases,it is neesry for the JRs to tion of RaihvayWorks' Union UR Sorcn).After reservetheir own decision-makingright on impor- restucturng in 1987, separate workers' unions tant investment projects or involvementin such wer organizedin the newJRswith JRSorenas the projects in order to avoid becoming, in effoct, suprem body. Theseunions andJRs signedjoint anotherJNR. declarions and establishedscable labor-manage- Uxion Problms and Eliminatie f Union meat rlationships. TIhy are legay constituted unions,which are not subject to the speciallabor Rqukdm lawfor publiccorporations,and arc giventhe right 3.122 AtdhetimeJNRwasestablishedinl949, to strikeaswellasthe rightofollectivebaainin a speciallabor law (PublicCorporaton and Labor Each union negotiatedwiththe managementsepa- Japanese National RailwaysPrivatization StuldyII 51 racelyand consequendlythe rate of pay risesbegan non (JR Rengo)in May 1992.In theseunions, dte tO be differentiatedfrom the first ycar of 1987. old etsurm group was dominant. Consequendy, 3.125 In 1990 confrontation over hegemony the old Doro group formed additional minority withinJRSorenbeween theold DoroandTctsuro unions inJRWcst,JRCntra andJRKyushu. On unionscane to the surface.InJuly 1991 theunion the other hand, thosewho are against the current ofJR WestsecededfromJRSoren and the union of leadershipoftheold Doro group in the union ofJR JRCentralfollowedsuitinNovember.Theunions East formed a minority Union and joined JR ofJR Shikolakand JR Kyushualso seceded around Rengoin January 1994. Thus, the rivalrybetween the same time. They formed another supreme JR Soren and JR Rengo has been incrcasingly body, theJapan RailwayTradeUnionsConfedera- intensifiued IV. MarketStructure of Transportionand ComparisonofJRs and Private Railways

Market Structure of RailroadSector in Japan areas, electrificationin this region is not urgndy Railbad System mjuapan requirecLAmong the rest of the regions,the share of electrificationin he Shikokuisland is far behind Radoad RouteLexgth by R the averageof the other regions.Since the privati- 4.1 TheiloadsysteminJapan hasfollowing zationof dhe JNR, the JR Sbikoku has constandy characteristics Firsdy, passenger traffic, not the and cagedyundertaken cIectification projects tO freight traffic,is the main busincsssince long timc be coampetitiveagainst road traffic. (in FY 1991 400 billion passenger-kmversus 27 4.3 Double tracdingis also an important fac- billion ton-kin). Secondly,in addition to the sub- tor for speed up and railroadcapacity. The JRs in ways, the privaterailways are concentratetdmosdy the Honshu (mainland),wherethe largevolumeof (about fifiypercent of the total route lengt of the trafficis expected,have a rclativelyhigh share of private railways) in such metropolhtan areas as double trning in terms of route lengd. On the Tokyo, Osaka, and Nagoya, and thus operate in contrary, since the railroadin the three small is- the most populated areas. On the contrary, the landsarelocated inthelessdenselypopulatedareas, JRshavetheirtrackinfrastructuresinborhpopu- the JRs in these regionshave a low shareof double lated metropolitan area (only about eight per- tracking,as low as less than one quarter. cent of the total routc length of the JRs) and also Riwra, Traffc CAaraceriwics intercity arcas (trunk lines). Thirdly, the rail- road in the three small islandsis overwhelmingly Passenger Trfic y DiswAwe dominated by the JRs (86% of the total routc 4.4 AccorDingEotheTabeIV-3fo)llowinutnv- length, whereas 71% in the Honshu), where few elerson the bullet traminand the other trnk lines large cities are located and the population is less withouta commuterpasshavealongtrip disranceon dense. Among the three islands, railroad in awerg and thuscan be groupedasadifkrnr typeof Hoklkaido is located in the most sparse area in customersfrom the short distance traveles The terms of squarc Iclometcr and of population. marketstructureof raoad injapanwill bereviwed Elemficdaionand Double Tracking hereaferbytypeofthemarket, ice,thelongdistance passengermarket, the commuterpassenger market 4.2 The gographyofJapan ischaraceisedby in metropolitanareas, and thBefreight market large mountainousarea and scattered metropoli- tans on the coast. Thus electrificationof the rail- LongD PaxengerMirket road, whih can provide a stronger traction or 4.5 Table IY4 followingshows that the ma- acceaion power oftrains. isan important factr jority of long distance passengersby rairoad ar for speedup and consequendyfor the compeitive- businesstraversm For te privmtepuposes of trav, ness of the railroad. Since the Holdcaidoisland travrbycarismuchmorepopular.Thusthedemand occupiesless mountainous and frwer metropolitan for long distance travel by railroadshould bc af- Japanese National Railways Privatization Study II 53

TabbIV-1: Rout Longthof Janem Railrod in March1993 (An, %, kwOO qnL, peaonk) Density Desty iRs PR? Total (Amre)** (Pop)** Three Islands 5,584.8 85.6 942.1 14.4 6,526.9 45.2 3,545 Hokkaido Island 2,628.3 92.3 218.6 7.7 2,846.9 34.1 1,984 Shikoku Iland 855.8 82.5 181.9 17.5 1,037.7 552 4,035 Kyushu Island 2,100.7 79.5 541.6 20.5 2,642.3 62.7 5,034 Honshu (Mainland) 14,555.8 69.9 6,2812 30.1 20,837.0 90.2 4,784 East 7,502.0 71.6 2,979.6 28.4 10,481.6 80.7 5,157 Center 1,994.7 58.1 1,441.4 41.9 3,436.1 117.3 4,157 West 5,059.1 73.1 1,60.2 26.9 6,919.3 96.4 4,529 Total 20,140.6 73.6 7,223.3 26.4 27,363.9 72.9 4,488 Metro Area*" 1,631.0 30.9 3,639.2 69.1 ,,270.2 n.a na. Three Metros 1,587.4 37.3 2,664.4 62.7 4,251.8 272.0 11,722 Source InsmuteofTcunsporamion Economics, TashiK*nx Npe (Menpeftan TdN/fAnZud. MinisnyofTanspon. Toimu Yearm(RatbvaALma) and Tesmi Yae TeokNnpe f(RavoadTrak StauAnwa. Others. * PRsindude Eidan (Teiw RapidTrnsitAuthority) and other publidy owned nlways.and allocad in each rqpon by the arm ofthe Tranportaon Administaion whichis notwhoiy conpaibewiihthe anra oered by the conspondingJ - ArmDensity ida utelength (Ian) per thousand squar meter. - PopulationDensity idcates peson per rourclength (kin). -Meo Aea sents suchmajor metropolitan areas as Tokyo, Osaka. Nagoya, and otheraty areaswith a populationof over milion (Datafram Teado Yus TokaNeio). Thre Me ar Tokyo,Osak andNagoya meropolitan aras (Data in March1990 fiom TodnKi Neap). Note:The nexrright olumns ofJRsand PRs epreat the percent sharesofJRs and PRsin the totl roumelength. Am ofthesurfc spaceand the population are nor wholly compatible with the areacovered by eachJR. Populadonfigurc is an cstimat asof October1991 based on the nationalcnsu

TRabWIV-2: EBetrIfctIon andDouble TrMck kglorJRs (Percent) Electrificaton Double Tracing April 1987 March 1993 April1987 Mawh 1993 Holdcaido 9.5 16.5 18.7 25.5 Shikoku 6.4 27.5 3.9 6.1 Kyushu 45.2 52.0 21.7 24.4 East (65.4) 69.2 (703) 73.6 (47.7) 53.4 (49.4) 55.1 Center (67.0) 76.1 (62.1) 72.6 (46.0) 60.9 (42.5) 58.4 West (50.0) 52.5 (53.2) 59.1 (32.1) 40.3 (36.4) 44.5 Tota (47.5) 52.5 (53.5) 58.2 (33.9) 40.1 (37.3) 43.6 Source Ministy of TRISrport TemsudoYorn (RailmadAlmanac). Figre in () excludeShinkansn ties. * DoubleTracking indudes double nadcing and morethan doublctrading.

TableIV-3 Aerage TripDisance by Pail(FY 1992) (krntrp) AverCa Pase Non-Pas JRs TrUnk 56.2 23.8 101.1 Bule Train 264.2 87.6 278.5 Others 32.1 22.9 48.4 Metopolitan 14.6 15.9 12.2 Other Local 173 13.2 25.2 PRs M Opolitan 11.2 11.9 9.9 Subway 6.7 7.5 5.5 Other 13.7 14.2 12.8 Othr Local 9.0 8.5 9.5 SourwMinity ofTnspom Tesudo YusoTokei Ncnpo (Riw TrafficS Annual). Note OPassrprsents commuterpass pas_nrs Tunk reps al thetunkc lines outside h metropolitanarea 54 Japanese National RailwaysPrivatization Study II feccedmorc significandyby die levelof business haveashareofasmuchasfortypercentintrennsof activitiesrather than by that of householdsector total tonnage. activities. 4.9 Since Japan is surrounded by ocean and 4.6 In this categoryof travl, air is the strong many industrialsites are locateddose to a port or compctitor for railroad. The competitiveadvan- coastal area, ocean shipping has also been fairly tagebetwcentheairandthemailroadcanbesumma- common modc for long distancefreight and has rized as follows; about fortyfivc percent of the totalfreight trafficin 1. The railroadis competitive,if the diffirr- terms of ton-km. Most of the wrmainingfive per- ence of the traveltime requiredby railroad cent is transportedby railroad,and air freight has and by air isless than 150 minutes(eg, only a nominalshare. three hours forty minutes by bullet train, 4.10 Amongthe railroadfieight traffic,freight and one hour and twenty minutesby air). tcafficbybulklcargoesandcontainecrshavediffercnc 2. The air is competitivc,if the differenceof characteristics.Firsdy, the averagedistance oftraf- the travd time required by railroad and by fic for the bulk cargocs(207 km) is much shorter air is largerthan 200 minutes dtan dtat for the containers(924 kin). Secondly, PassengerMarketin MeurmpoEtanAa lines of commoditiestrnsported arc different About30% ofthetotal tonnagebythebulicargocs 4.7 Passengersin the metropolitanarea are is for petroleum and peroleum products, about doninated by passengerswith commuting pur- 20%forcement,andabouil5%forlimestone,and poses,ie., businessmen and womcnand students. thus dhesethree commodities ofbasicmacerials add Thc share of commuter pass passcngersis about up to two thirdsof the total tonnage.These com- two thirds both for the JRs and for the rmilroadas modines are transportedby spcial cargoesmosdy a whole Although the purpose of trsav by the owucd by shippers.On the other hand, commodi- pssengerswithacommutingpassisnotnecesrily ties pored by the containersare much more its genuinepurpose, a Laureshare of the passngrs varied and valueadded, in other words,procssed with a commutingpass ensures that the railroad oodsorconsumergoods;,about20%for chemical demand in the metropolitan area can bc muck products. about 16% for food products,and about morc stable than that for long distance travelby 12%formtileandag a ralproducts rpecte. rallroad. 4.11 Compedtionandsharesamongthemodes Frer*htMarkt offiighttaffcinJapanaresummarizedasfolUows; 4.8 Road(truck)istheleadingnodeoffreight 1. Truck (frcighttaffic by road) is the domi- trafic in Japan, which has a share of over fifiy nant (over50%) mode of freight trffic with percentin wmnsofton-km. Suicethe freightu-dic a distanceof less than 500 kin, but with a by mad is mosdyconcenuratd in a short distance distanceof over 1,000km, the shareoftrck is trffic,itsshareintemsoftonnagereachesasmuch decreasedto only about 20 percent. In other asnnetyonepaecetAthoughcommoditiesmans- words,88 percentof the wotaltonnage by tucdk ported byroadhavealargevaricty, gmtlandwaste is a trffic with a shorterdistnce than 1001km.

TabWIV-4 P'upow of Long Ditnc Tre hI OneAutumn Day (Numberof Travelst/erc Share ByAir ByRain ByCar ByOthers Business 28.3 10.4 36.8 50.6 2.2 Sightsecing 42.7 n.a. 5.0 92.6 n.a. OtherPrivate 13.6 na. 12.7 80A n.a. Others 2.8 8.1 37.6 48.1 6.2 Not Spedfied 12.6 na. na. n.a. n.a. Sour: Moroboshi.anRsRyoKarsuidonoSogoHaadmniKansuruChosa(R hontheTmnkTrafficforNP2egeu). JapanPort & HarborAssodation. Kowan. Mard 1993 Peiodof Rearh-:FY 1990 and 1991 Cdlumn'Shae indits theshare of tavdelswith ach purposeof ta in thewtol number of tmver ColumnaByAire. yP.Rail and so on indicat h sharoftvers byeachtaffic mode in thespcific puxposofveL 'n.a. inthe table indicats tht datais not available or thatthe figureis negligible. Japanese National RailwaysPrivatzation Study II 55 2. For a tafficwitha distance of over 500 Ian, Since Hokkaido is a nordhen island isolated by the total trafficby railroadand marine ex- road network,the sbare of road is lss than nine ceedsthat by road in termsof tonnage. percent.Although Niigaa is locatedadjacent to a Among railroadand marinc, trafficby ma- big markaeof Kanwoarea, dte distance between rine forms is alwaysmuch larger than that by dthse two regionsby ocean route is much longer railroad,as much as ten timesof that hy than that by railroad, and thus the marine traffic railroad.Thus railroadfor the long distance between these two regions has only a nominal freight traffic,which is dominated by con- share. Those geographicconditions pardy deter- tainers, is no morc than a marginalcarrier minc the advantageof the railtoadtraffic. which has a competitivenessonly for the 4.13 In the case of commoditiestransported a inland taffic long-distnce (exceptfor occan shipping),raroad 4.12 By origin and destination of traffic, rail- container freight has a majority share only for roadhasa rcladvelyhighersharein the trafficto and textile products. AU the other commoditytrans- from Hokkaido, Niigata, and Tohoku regions. ported isdominatedbyaroadrraflc.Especiallyfor

TablIV-5: She ofCommuter Pas Paenr InMeofolIan Areas In FY19 (Perentage) All RailwaYs JiRs Tokyo Area 65.4 66.0 NagoyaArca 56.6 43.6 OsakcaArea 61.2 63.6 Three Area Total 63.8 64.6 Sourc: Instiute ofTranspormion Economics, Toshi Kis Nenpo(Mtepowp T :ftAsiwQ.

TaWeIV.- Fre Trafc Shae by Modeand by Disan (FY1991) (Thouatnd kmI) JR Fright Dstance (cm) Totl Con.ai Bulk Tndek Marine Totl -100 15,421 10S 15,313 5,395,351 135,387 5,546,159 101-200 11,986 273 11,713 336,933 75,987 424,906 201- 300 5,921 433 5,488 135,407 53,389 194,717 301 - 400 2,978 1,439 1,539 78,658 53,424 135,00 401- 500 2,276 1,249 1,027 50,582 48,276 101,134 501- 1,000 10,893 8,846 2,047 96,916 124,599 232,408 1,001- 8,826 8,213 613 19,718 68,965 97,509 Total 58,301 20,561 37,740 6,113,565 560,27 6,731,893 Ave.Dist. 465 924 2)7 45 434 81 Sourcelnsumte o TranspornaionEconominc Unyu t KeuaiCrnrsporction & Economy),September 1993. Koum Kyok (ransport Asscion). KorsuNenpo (raffic Annua). TableIV.7: Long Ditnce FreightTraffic by Road and Rai and by Commodty (at InFY 19) (Thousandtn, %) Total Tru C Shre of coniner Mining Products 122 65 57 46.7 Wood Products 197 160 37 18.8 AgricultumralProducts 1,027 780 247 24.1 Chemicalproduct 1,047 648 399 38.1 MachineryProducs 1,292 1,159 133 103 FoodProducts 904 576 328 363 TextileProducts 299 80 219 73.2 Other Products 1,909 1,783 126 6.6 Others 2,598 2,123 475 183 Totl 9,395 7.374 2,021 21.5 Souro=M:nsyotTrnspo= Kamotu nik KyudoCi. 1yokyakuchuii KyudoUlzosa (me Arad 1 TrA Statsics. Dam(orTruck" is arbit.raycalcaud basd on tde I bet regions.Od oa=uri o and strd, paperand pulp, and so on. 56 Japanese National RailwaysPrivatization Study II machineryproducts and other products, the rail- in nominal term during this period,by abolishing road has only a nominal share.Those may be the the Passenger'sTax of 10% (nct of about 6.5% potential categoriesof commoditiesfor the modal decrease together with the introduction of the shift from the road traffic to the railroad traffic. ConsumptionTaxof3Yo)sinceApril 1989 and by Neverthelesssuchservicequalitiesasfrequency and introducingvariousdiscountticketssinceFY 1990, flexibilityof trafficschedule arc important consid- resultingin the decreaseof about 17% in real term erations for die choiceof mode of traffic. comparedto that in ten yearsago. Therefore travel by Pgsengfr Tr4;c air has alsoincreased significandy by 9.5 percentper year.Tavel byrailroadhas also achieved a goodresult Dmand ELarmicivof Traffic of 4.9 percentgowth per yearduring this period. 4.14 Passengertraffic demand in Japan shows 4.17 Afierthe strong economicgrowth period, several fiatures. Firsdy the demand during the the inaeaseoffrectimcand impetusforleisure has secondhalfof l980's grew muchmore rapidlythan alsobecn saturatedand thus the long distancetravl the generaleconomy, the demand elasticityto the has recendystagnated. GNP reachesas high as twu, which is quite high PassengerTrafi- in Mer*p.iia Area comparedto those of the other perods. The odter 4.18 During the secondhalfof 1980's, passen- feantureis the stable elasticityofdemand for the PRs gm Durinthe merop t ha als, bem throughout the period. Since most of the PRs gei traffic in the metropontan area has aLsobeen operatein the metropolitanareas, the demand for ncreasrngwint a fairly strong pace of 3.7 percent the PRs ismosdy concentratedon die commuting per year, if not so strong as the long dtance traveL and daily traffic which may not fluctuate much Among the modes of traffic.travel by privatecars throughout the period. On the contrary,since dte haveincsedwithamarvelouspaceof7.0percent JRsrelymuchmoreonthelongdistancetravd, the per ycar, supportedby the increaseof the owner- demandfortheJRs is muchmore vulnerable to the ship of private cars and the decreaseof gasoline change of economy, busincss activities,and the prces. Exceptfor the privatecars, the JRs achieved people'staste for the leisure time. The demand for a beter performancethan any othe competitors. the bullettrains seemsto be a typical examplefor 4.19 Passenger-kmdata also support the good the long distancetravd, vevnthough itS opeatio-i resultsfor theJREsduring this period. Owingto the thsuklmay bc ovtrvated bevent of its oDnstnr cffortsby -he JRs and thus the improvemcntof development throughout the period. Other JRs thr mpeivc amongdie 2ailzoadtaffic, tan the bullet trains have reoDveredonly to the rhe JP?shave achievedsrrong successduring FY vedofthePRsevenduringthisprosperousperiod. 1987 to 1992 ir the metropolitan market, com- pared bothto the PRs and to the subwap. Lhe LongDisncer PaexngerTraffie successby the JRs is attributable both to the in- 4.15 During the second half of 1980's, thc crase ofnumberof travelersand to the averagetrp Japancse cconomy has grown rapidly. Both the distance by avelers. Duc to the spread of the businesssector and the householdsector have ex- metropoltan ara and dus he cxpansionof com- panded their expenditure significandy.The real mutng distance,the averagctrip dist b) met- prvate final consumptionexpenditure Cm the Na- ropolitanrcsidents has been expandedduring this tionalAccounts)has grownby44 percentperyear pcriod.In this regard, the JRs havc capmred thais on avragc. In addition, two days off a w has expansLioneffect most successfully. becomemuchpopularbusinesscustomandcthefree Fr8igbt Traffic time has increasedfor many Japancse. Owing to 4.20 FreighttfficinJapanhasbeenchraczer- these income and time factors,the long distance ized by a huge share,about fifiy percnt, of taffic travelduring the penod has incased rapidiy. by marinc forms dtroughout the period, owingto 4.16 Among three mnajormodes of traffic, air its geographiccondition. WViin the r and road have hived a remarable growth The iialfofthc ramfic,sharchasshted fiom the ailroad successof road taffic is attributed to 1) develop- to road traffic. ment of higways, 2) increaseof the ownershipof private cas, and 3) decase of dte real gasoline LoxgTerm Reviewon AheFrget Trafl prices,on top ofthos: faictorsmentioned abovc. As 4.21 Duringth period ofearl1980's,eventhe for air, the averageair fire has been decreasedeven total ficight taffic declined and truck taffic in- Japanese National Railways Privatization Study II 57

TableIV-8: Passenger Traffic Demand (Pasengerkin) Elasticity of PrivateFnal Consumption Expenditures FY Total Railroad JRs- Bullet Othcr JRs' PRs '75 -'80 0.525 -0.169 -0.660 -1.587 -0.404 0.611 '80-'85 0.669 0.323 0.150 1.221 -0.221 0.581 '85 -'90 1.877 0.820 0.936 1.637 0.630 0.635 DataSourc Econo mic PlanningAgency, Annual Kepo rc on National Accouncs. MinisqyotTranspor .WhaitcPaper onTransporr. * Tomilindicats all modesof traffic. `JRs before1987 isJNR.

TableIV-9: Traffic Distance of Passenger (BillionPassenger-kmn, cars per day) 1987 1988 1989 1990 1991 1992 1992/1987 Rail 95.3 103.2 103.9 112.4 116.4 114.8 1.205 Air 38.5 41.1 47.1 51.6 55.4 56.7 1.473 Highway' n.a. 54.7 n.a. 91.7 n.a. n.a. n.a. Highw2y"- 3,525 3,894 4,220 4,550 4,701 4,917 1395 Soure: Vata tor Kailarc 'TrunE Lineoutside ot metropoimanarca crafficdam exccpt for commuter passpassengers. Minisyot Transport, Tcrsudo YusoTokei Ncnpo (RailwayTcafficStatisticsAnnual). Daa forAin KowsuKyokai (TransportAssociadon), KonsuNenkan (TrafficAnnual). Daa for Highway are in billion passenger-kmdinduding riversoftrudcs Miyaokand Kato, Heisi 2 Nendo Doro Kotsu Sensasu ni Miru Kosoku Doro no Riyo Jitmai,Express Highway Racah Foundltion of Japan, Kosoku Doro to Jidosh (Expresswaysand Automobies), May 1990. Data for Wighway are in as per day in highwas and toll roads. Japan Road Public Corportion.

TaebIV-10: Metropolitan Traffic by Mode (MillionPasengers) 1985 1986 1987 1988 1989 1990 19011985 Railroad 16,523 16,966 17,311 18,047 18,436 18,974 1.148 JRs 5,556 5,730 5,907 6,213 6,307 6,608 1.189 PRs 7,318 7,483 7,620 7,799 7,975 8,089 1.105 Subways 3,649 3,753 3,784 4,035 4,154 4,277 1.172 TraM 113 112 108 110 109 110 0.973 Bus 3,576 3,558 3,567 3,583 3,577 3,605 1.008 Taxi 1,482 1,504 1,524 1,544 1,511 1,412 0.952 Car 9,689 9,863 11,015 12,197 13,455 13,590 1.403 Totd 31,384 32,002 33,525 35,481 37,089 37,691 1.201 Source:Insanie of Transportation Economics. Tuhi Koau Nenpo (Menwpditsn TnfflcAwaO). Note Merropolianhere indicaesTokyo. Osa, and Nagoyametropoltan areas.

TabeIV-1 1: Railrod Trafficin Metropolitan*Ares (Billion Passenger-kn, kM per Trip) Traffic Tnp Distance 1987 1992 1992/1987 1987 1992 JRs 70.8 88.5 1249 14.0 14.6 PRs 108.3 116.9 1.079 13.8 13.7 Subways 28.2 31.6 1.123 6.7 6.7 Totl 207.3 237.0 1.143 12.1 123 SourceMinistzy otFTransport, Teasudo Yuso Tokei Ncnpo (Railay Trattic Sursuc Annual). Note *Merolipoachendieatscsudch big ityareasasTokyo.Osa Nago2y.andodiercitieswidia population ofoveronemillion.

TableiV-12 Share of FreightTrffic by Mode (Millionton-km %) FY Total JR PRs Truck Maxinc Air 1965 186,346 30.3 0.5 26.0 43.3 0.0 1970 350,656 17.8 0.3 38.8 43.1 0.0 1975 360,490 12.8 0.2 36.0 50.9 0.0 1980 438,792 8.4 0.2 40.8 50.6 0.1 1985 434,160 4.9 0.1 47.4 47.4 0.1 1990 546,785 4.9 0.1 50.2 44.7 0.1 Source:Miniscy of Transport.Whim Paperon Transporm 58 Japanese National RailwaysPrivatization Study II creasedless than the tota economicactivities. The November1986 and all the railroadfreight service reasonfor this weakdemand can be attributed to, has boenconverted into the direa services.And as I) outsourcingof bulky materialsfrom abroad, 2) the railroadhas concentratedits investmcntstrat- product shift to lighter,smaller one with a higher egyoncontainersandintermodal f2cilities, most of valucadded,3) efficiencyimprovementofdistribu- the trafficinrease for the railroadfreight has come tion and deliverysystem, and 4) increasingshare of from the trffic by containers,not by bulk cargoes. serviceindustry in the whole economicacrivities. 4.25 Neverthelessthe competitivenessof raa- Although many of these factorsseems to be effec- roadfrcighttraffichasimprovedbyuneven amounts tive throughout the recentyears, the elasticityfig- among regions. ure during the second half of 1980's does not 4.26 Intermsoforiginoffreighctraffic,railroad confirmitand the freighttraffic hasstarred to jump has achievedan imnpressiveincrease in icscompeti- again in the second half of 1980's. tivenessfrom Shikoku,owing to the inauguration 4.22 Railroad ficight traffic in Japan has con- of the railtroadbridge between the Shikokuisland standy reduced itsvolume of transport and also its and the Honshu (SetoOhashi) in April 1988.Next shar among modesof trafficup to the firsthalf of largestsuccessforrailroadisthetrafficfiomNiigata, 1980's.BesidesjNRmanagementproblems,changes for which the total traffichas decreased,whcreas in linesofcoonmodities and ofdistributionsytem in the railroad traffic has kept increasing.The de- Japan caused the dedine of raiLrad fisight traffic creaseoffreighi rafficfrom Niigaa isattributable duringthisperiod. Plasticizauon and dWesification to the trafficshift fromNiigata to Hoklriku which of commotities providesadvantage for the traffic is adjacnt to Niigata, owing to the start in fizll by road and by container.As for the distribution serviceof Hokariku Highwayin July 1988 which system,high qualityof servicehas been required to connectsthe Hokuriku areawith Chubu and Kinki accordwith a"Just-on-Tine systemofproducer. areas.On the contrary, this resultsin the loss of These changeshave promoted downsizingof or- railroad competitiveness in thc traffic from ders and prompmessand ceraintyofdelivery, and Hokuriku. consequendyincurredtheeclinceofrailmadfreight. 4.27 In wtrs of destinationof affc, Shiloku AreentDevelopment of e Rroad Freght is alsothe big winner for railroadfieighr. Same as pment of the Railroad Freight the trafficby orign Niigatais the secondsuccess -ompetitivmw for railroad. Contary to the traffic by origin, 4.23 During the second half of 1980's, total railroadhasg ineditscompetitivenessinthefreight fieight traffic increasedfaster than the economic rafficto Hokkaido,owning to the inaugurtion of growth. Ralroad freight traffic,which had been the "Seilan Tunnel' in March 1988 which con- constanty declining, had started incrasing once nects the Hokkaido islandwith the Honshu. againby an almostparalle pacewith the real GNP growdt during thisperiod. On top ofthe favorable Comparison of Permance between JRs and economicsituation, the shortagcof truck drivers Pn Raiwas and the congestionon road had encouragedsomc 4.28 Few people dispute the successof JNR's freight trafficshifting from road to railroad. restructuring.Efficiency gains ofnew railwayenti- 4.24 In addition, in order to accord with the tics compard with the old JNR are analysedin changesin the businessenvironment, mashaling detail in the World Bank publication "Japancse yards were abandoned and instead traffic centcrs NationalRailways Privatization Study." Howcvcr, were establishedin February 1984. Furthermore the question remainsas for to what extentJRs are the trafficcenters have also been abandoned since operaing efficientlyin comparison with private

TabeIV-13 Freight Traffic Demnd (Tonrn)Elastidty of RealGNP Total JRFreight Truk 70-'75 0.139 -1.766 -0.243 75- '80 0.902 -1.323 1.390 '80 - '85 -0.063 -4.192 0.771 '85 - '90 1.019 0.985 1.232 Sources:Economic Planning Agency. Amwa RepmoenNaziwAceun. Minisy ofTransport. 174ike Paper on Transot TableIV-14: Competitive Indicator of RailroadFreight by Area (1987to 1991) 1

I ~~~~Destiation of Traffc Area Hokkaido Tohoku Kanto Niigata Hokuriku Chubu Kinki Chugoku Shikoku | Kyushu Hokkaido -23 x:?el*iss<4' "8 -57 Tohoku > 1 -14 j -59 -29|v 4 <37I -54 Origiii Kanto "';< s,g < >.?7 of ki .>- 11 Traffic Hokuriku 3 I Chubu__3 Kinki __ : Cugoku|? -29 1__I8 Shikoku _ mlom

. r~otal lil li N}-41 -81 I M flm m DataSource: Ministry of Transport,Kamiotsui Cfiiiki Ryuido CElosa, RyoAaku ClilikiRyudo C/losa (finterArea Freighlt anid Pa.ssenger Traffic Statistics). Notes: * Competitivenessindicator is the differencebetween the increaserares of freight trafficin tonnagc by railroad and that by the total modesof traffic. Yearsof data used are FY 1987 and 1991. * Traffic within these areas is neglected,because these distancesare too short to use the railroadfreight. * Competitive irndicatorsugests _: ~~morethar 50

" ^'t*th |betwveen Oand 50 :lessthan O

D 60 Japanese National RailwaysPrivatization Study II railwaycompanics. In Japan, in addition to J Rs, many smalllines were constructed in localareas. there isalarge numberof private railwaycompa- By 1905, the routc length of private railways nies.AAmongtheseprivatecompanies,thelargest reached 70% of total railway route length in 15 companies (in terms of capital, route length Japan (private railways 5,231 km. public tail- and passengcrs) are refcrred to as the Large ways 2,413 km). Private Railways(Large PRs). The achievement 4.32 During the recessionof the early 1900s of management on a par with the Large PRswas following the Russo-JapaneseWar, these small one of th: guidelines of the JNR privatization. private railwaysfaced bankruptcy. At chesame In this section, following a brief outline of the timethcJapanesemilitatybecameconvincedofthe Large PRs,we consider the effcctsofJNR priva- railway'sstrategic importance and lobbiedstrongly tizadon by comparing the JRs to the Large PRs. for the nationalization.In 1906, a railwaynation- The data in 1991, the year when detailed data is alizationact waspassed due to the financialprob- available,is used in this chaptcr unless otherwise lems of small private railways and military specified. Later, in Chapter V, an attempt is considerationsin spite of the opposition of many made toestablish the methodology ofeconomic private operators. Afer that, the main intercity approach using regressionanalysis on this topic, railwayswere operated by state-owned railways, talding into account the network condition and managedby the Ministryof Railways,and privatc serviceoutput. railwayswere limited to urban and local railway Lae Privae Rawys transportation. 4.29 I'henamesofdhel5LargePRsbyrcgionar,-- 4.33 During the Taisho era (from 1912 to 4.29 Thenamesofthe1LanrgeiPRsbyT gionarc 1926),as thepopulationgrewrapidlyinthecenters T4,reSa (8 RailwayCompanes) Tobu Rail- and suburbs of largc metropolitan areas such as Kway,Seibu Railway,Kcisei Electric Railway, Tokyo and Osaka, the demand of private railways Raiway,Tokyu Corp., Kehin ElcctricEx- increasedastheagentofinnercitycommutertrans- prcssRailway, TokyugCorp.i Keihwayn Elecportation. However,the recessionin thebeginning press Rai dwwaySa ompa Railia k ^ofthocShowaera(from 1926)and thespreadofbus ponAra (l RailwayCompanic): Kinki, Nip- transportationbrought about a rcorganizationof pon Railway,Nank CorpE.lecHanscin Ela - privatzrailvays. Large private raihwaystook over

* Raihvwa - smallprivaterailways.The imainingrailwaycom- tl C) N paniesfound their way into such other businesses NagoyaArea (IR as rcal estatcand tourism. Railroad 4.34 Aftcr Wodd War II, transportation de- Fukka 1 RailwayCompanies): Nishi-Nippon and in large citis ineased sharply. Private railwaysreorganized and 14 larg private railway Historyofthe PRs in Jap companiescame into existence (the number of 4.30 The first privaterailway (Nihon Tetsudo) large PRs became 15 in 1990,when Sagami Rail- openedbetween Ueno and Kumagawain 1883,11 way, dassifiedas smallprivate railways until then, yearsafter the inauguraton ofthe firststate-owned wasadded to the largeprivate railways). With high railway, Japanese National Railway.The Meiji cconomicgrowth in the 1950sand 1960s,popula- government,which wanted to 'catch up' with tion became concentratedin large mtropolitan industrialcountries in Europe and America,pro- areas.The increasein populationin the suburbsof moted the constructionoftransportation nerwork. large metropolitan areas made private railways Since the government lacked sufficientfinancial indispensableas high-speedmass transportation resourcesto establishatransportation network, and system.In addition, the population increasegave becausc ofits policyavoiding dependence on finds privaterailwaycompanies opportunities to partici- fromoverseas, it encouragedthe privatcsector to pate in side-businessessuch as real estatedevlop- invcst in railwys. Followingthe successof Nihon ment, becausehousing demand increased along Tetsudo,pnvate cntrepurs consuaced raidways privaterailway lincs and privatc railwayssold off one afier another. It is noticeablethat the private the land and residenceswhich they developed. railwaysat that time constructedintercity lines. They also opened departmentstores and amuse- 4.31 Afterthl 1892recession,thcsecondboom mcntfci;itiesinftontofthesations,whichboosred of private railway construction begn This time the advantagesof privaterailway lines. Chart IV-1 Railway Network in Tokyo Area

k I LI) Lime) i z

Ut

.s.:!=P. ]cu.P LI

Tbra.peu1aiImAm.

JR.

- NewTrampod System ChartIV-2 - --- - Railway Network In Osaka Area i

FukuchlyamLleine , -- -L

h li

.0 ~~ ~ ~ ~ ~ ~ ~ ~~

hm I , I _ _ .h ______1 1 1__

< rieshimelo z * * * | ~~~~~~~~-New Tlportatm Systeal k a hI7~~~~~~~~~~~~~~~~~I

| _ _ & , _ | ( ~~~~~~~~~~~~~~~~~~~~~------kKatarlkl |i|e_ JapaneseNational RailwaysPriviadon Study II 63 4.35 Aftertheoil crisisofthemid 1970sandthe crate.The consolidatedsale ofthe Large PRsis two cndof higheconomic grwth, thesituationchanged to three timesas much as dtat of parent companies Populationgrowth in metropolitanareas slowed. basedontheconsolidatedfinancialsutements,and Furthermorc,with the end of the land develop- wouldbemoredanthatifindudinggroupcompa- ment boom,private railways have not camed prof- nies not covered in consolidatedfinancial state- its fiom real estat developmentas much as they menhs. once did. 4.39 Fares are regulatedby the care regulatory 4.36 In 1987 JapaneseNational Railwayswas system,leading to rigidityof revenue flow in the privatizedandonceagainthcprivatesectorcameto railway busincss. Its profitability is strongly af- operate most of the railwaysin Japan. fectedbythecimingandsizeoffareincreases.In the GbtaofPPs context of such negative aspects of the railway business,the diversifiedbusiness sector givesthe 4.37 The Large PRs mainly transport people LargePRs stable arnings.For example, in the case commuting to work and schoolin the thr main of developmentof such areas along the railway metropolitanareas (the Tokyo,Osaka and Nagoya tracks through the selling off of real estate, the areas)and Fukuoka,where the populationdensity companiesthmselves have been able to regulate is high. In the three main metropolitanareas, the the amountofland and residencessold, and hence share of passngers transportedby the privaterail- the flow of revenueand profit. Many of the Large waysis greaterthan that of the jRs. As the mnor PRs increasethe sellingoff of land when railway transporter unth largecia, the LargePRs have businessis poor, and restrict it when conditions mintainerelarivelysteadyprofits TheLargePRs improve. Thus ticy hae been able to ensure a receiveno subsidyfor their railwayopertion, steady flow of profit availablefor dividends over 4.38 The LargePRs have acively moved into eachfiscal year. Howver, becausethe supplements many other areasof businessbesides raivays. The that can be obtained from the real estatebusiness amounc of rcvenueearned by such activitiesaver- ar limited,their profitabilityhas begun to dedine gecs50% of total revenuc.These divasifid busi- recently. ness activitiesinclude bus opceraion,housing and 4.40 As a resultofthe lack of availableland and eal estatedevopment, the rental of office build- increasingconstuction COStSin the large dctiesthe ings, tourism and leisure, and other businesses. LargePRshavemadefwadditionsorexcensions to Divesifiedbusinessrdaredthoselivingdosetothe thir acr tain lines. Rather the Large PRs rilway is mutuallycmplementary to their railway concentate on the investmentforquadruplc tracc- business,shaping the growdt of the privat railwy ing and other capacity expansionsto relief the companies.The Large PRs also create separate congestionin metropolitanlines. companiesto implementtheir diversifiedpolicies, 4.41 The Lrg PRsarealllistedcompanies. In and the group of thcsc companiesforms conglom- addition, apart from the LargePRs, there are also

TableIV-15: Dhersified Business (y Billion) Honshu JRs LargePRs FiscalYear 1987 1991 1987 1991 Raiiway* Revenue 3,191 3,891 921 1,110 Profit 429 754 135 147 Bus Revenue 29 - 241 259 Profit -7 - 0 -6 Real Estate Revenuc - - 363 506 Profit - - 126 205 Others Revenue 23 93 280 408 Profit 17 16 5 4 Total Revenue 3,192 3,984 1,805 2,284 Profit 439 770 266 351 Source Minsuy of Tspor= Teiudo Toki Nenpo nnual Raiwas Siscics). - Note:Pro&fas opeatng prok - RailwayOperaingRevenuc Fare rvenue and other rvenue. nor includedsubsidies Railway Opeing Expenditure Labor. enelgy. epair. miselaneous pecnses.ox (not incuded corporactu and depreiation. 64 Japanese National RailwaysPrniatization Study II sveml other privte railwaycompanies listed on all Japaneseindustry (3.0%) and for all manufic- dh stock gehange. Thus the private railwaysin wuringindustry (4.3%). For dhcrailway business, Japanhavefirmbasis,asoneoftheleadingcompa- the ratio is roughly equivalentto the all industry nies, in their own businessareas averageat 3.2%. However,thc ratio for the othr FiwuidPaffornww busincssareas is high (7.6%), boosting the Large PRs' overallprofit ratio. Regardingthese ina 4.42 In fiscal 1991, the operatingrevenue for data for operating revenue and cxpenditurc of the LargePrivate Railway Compans cameto Y2.3 railroad sector are public Non-operating revenue trillion. Operating profit was Y351 billion, and and xpenditureofrailroadsctcorarecakulatedby current profit was V125billion. If wedisaggregare multiplyingtotal non-operatingrevenue and ex- the Y351billion operating profit, we seethat 1147 penditure, whichare reeased publidy in financial billion, 42% of tocal operating profit, is derived stareants,bytheratioofthcrailwayfixedasseuto fromtherailvaybusincss,while V2O5bilion, 58%, total fixedassets. comes from the real estae business.Thus profit 4.43 Profit and lossin the railwaybusiness arm from the real estatebusiness dearly exceedsthat of stronglyrlated to changesin fare prices. In fiscal the railwaybusiness. The ratio of curent profit to 1991, the operatingprofit ofthe railwayse':torwas totl saes is5.5%, higher than both th averagefor Y147 billion.However, due to the fact that 13 of

Tabl IV-16:Pri andLoss of fth Lare Priae RaiwayComnies (V'Dl Wn, Percent FisclYear 1987 1988 1989 1990 1991 1992 Operting Revenue 1,805 1,908 2,031 2,186 2,284 2,329 Optting Ependiture 1,540 1,614 1,738 1,864 1,932 1,990 Operating Profit/Loss 265 294 293 322 351 339 (Ratio of Profit to Revenue) (14.7) (15.4) (14.4) (14.7) (15.4) (14.6) Non-Operating Revenue 67 67 85 98 102 99 Non-Oprating Expendimre 231 231 241 297 329 318 Current ProfitLoss 101 130 136 123 125 119 Extraordinary -3 -17 -20 -9 4 -2 Revenuc/Expenditure Corporate Taxes 51 57 57 51 64 55 Net Profit/Loss 47 56 60 63 65 62 sour= MinistyofTransporcm Tersudo Tokei Nenpo (Annual Railways Sraidwcs). Tle V17 OperatingProft of the LargePRs: Railroad Sector and OtrSecors

Fiscal Year 1988 1989 1990 1991 1992 Railway 159 143 135 147 193 Others 135 151 188 204 146 Totl 294 293 322 351 339 SourcenMinistry of Tranorr. Tetsudo Tokci Nenpo (Annual Railways Saosdcs). 1992:Pne RailwayCompanies TabeIY-18: Scale of RailwayBusiness HonshuJRs LarePRs * Route-anm (cu) 14,545 2,870 Passenger (million persons) 8,199 8,009 Passenger-km (billi.n personkmn) 231.8 112.7 Traffic Share (pecent) 17.4 8.5 Siiu MinistryofTransport.. Tndo ThkNenp.(AwaLRAihuaSmtia), InstitnucofTrspomtionEconomincs5iSic MmwTasud '93 PUtFat Beek'93). HonshuJRs lbe toa of thethree jR companieslcatd on Honshu(mainland Japan) LargePRs: Th totl ofthe fifieen 1 pra rih y companies - Route-kn:One way.non-duplicatngtotl klometes Note: Thefigures are in 1991 JapaneseNational RaihwaysPrivatization Study II 65 the 15 LargePRs raised their prices in 1991, the increas in fares,the extraproceeds romwhicharc operating profit for fiscal 1992 jumped to '193 retainedfree ofcaxbythe companyconcrned and billion.As cosrstend to rise in the long term, the allocatedto such projects. On completion of the Large PR's incomc from their railway business project (completionmust be within 10 years)the tends to rcalizecontinuously declining profits in fitndisdismantledoveraperiodoftenyears,andin the periodsfollowing price adjustments.The rail- the end the moneygainedfrom thisretumed to the way business,strongly influenced by changs in users. While the funds allocatedare restricted to fareprices,is compensated for by otherareasofside 25* ofthe cost of the projectand must cvenrt-ally business,especially real estate. Thefollowingshows be returned to the users, it makes possible rhe the trend of operatingprofit of the LargePRs. In raising of interest free funds. At present, of the 1989 and 1991 when the operatingprofir of rail- LargePRs,5companies(Tobu,Scibu,Kcio,Odakyu road sector declined, operating profits of non- and Tokynu)are using this systea In FY 1994thc railroadsectors, almost all of whichis contributed amount of loansis expectedto increasefrom 25% by real estatc,incrcased. Howevcr, in 1992 when oftheprojectcostto 50%,andthe targetofprojects the operatingprofit of railroadsector rose,operat- to include new line construction. ing profitsof the non-railwayscaors decreased. 4.44 As mentdonedabove, in the long term, the stable railway business and the complementary 4.47 In this section performanceis compared naure ofthe other sidebusinesses ofthe LargePRs between the three Honshu JRs (JR East Japan, JR have meant they have bcen able tO rcalizestable Central Japan and JR West Japan) and the fifte dividendsofbeween 8-10% in eachperiod. How- Large PRs in order to assessthe effect of JNR!s ever, their profitabilityhas begun to dedine re- privatizaion. In the caseofJRs the three Honshu cendybecausethesupplementsthatcanbeobtained JRsare sdectedfromthe6JRpassengercompanies, from the rcal estatebusiness are limited. becausetheir operatingareas are mainly confined I C to thethreemajormetropolianrarcasandccoincides -wi"th the LargePRs. The sum of the dthreeHonshu 4.5 Asdiscussed above, the Large PRs havea JRs' damaandthesumofthefifteenLargePRs'data long historyas an imporant tansportation sysm are compared here. One diffilaty in maling a for peopleliving in the metropolitancities, devel- comparisonbetween the JRsand PRs is thar while oping in linewiththeir need The scaleand growth thcLargePRs are beavilyconcent;raedin the major of Japan'sLarge PRs is unique among the private metropolitan areas, the JRs operate many loald railwaysthroughout the world'slarge cities.How- train lines. In this context, there is an obvious ever, the Large PRs are presently faced with a differencin the trafficdensity of the two groups, dificult situation: Over-congestionduring peak making a closecompanison problematic. Because period is serious in mectropoPtanrailways, espe- of tiis, we analyze the diffirent nature of the ciaily in the Tokyo area. Many PRs in the Tokyo Honshu JRs and the LargePRs, while making a area carry twice as many passengersas they are comparisonof the performanceof the two groups designedfor. This problem requiresheavy invest- followingJNR's privatization. ment in order to relief the congestion and to 4.48 To comparc the producivity of the two quicker transportaion through such measuresas groups, simple average comparison medtod by quadruple tracking.However, due to the residen- actual figures is used in this chapter. Later, in tial crowdingin Tokyo, the acquisitionof land for ChaperV, economicapproachbyusingregression suchquadrupletrackinghasprovendifficult.More- analysisis sought, which considers the diflirence of over, the rise in land prices in the inner aiy has networkand service.As a rle, the period of com- mcant that immense sums of money to finance parison will cover the five years between JNR's such constructionare requird. privatization in 1987, and 1991 whn derailed 4.46 In order to promote quadrupl tracking figures(AnnualRailwaysStatistics)werepublished. and other such large-scaleinvestment by private Baic Figursfor the HonshufRi ad the Larg PBs railwaycompanies within the metropolitanarea, a new investment-promotingsystem was established 4.49 The total route-km of the Honshu JRs by the govemen in 1986. Under this yastem, comcsto 14,5451an,comparedto2,87Onmforthe constction ispariLly financedin advanceby an LargePRs The route-lm ofthe Iarg PRsissmall, 66 Japan: National RailwaysPrivatization Study II

TabbIV-19 Characbdtesic of Traffic Honshu JRs large PRs FY 1987 FY1991 FY1987 FY1991 AverageTraveling Distance (kln) 27.5 283 14.1 14.1 Trffic Densitry 35.4 43.5 100.2 1073 (thousandpersons per kmn) AveragcPassenger per Car 51.9 53.5 66.3 63.4 (personsper kIn) Sourc Minisrryof Trnsport, TersudoTokei Nenpo (AnnualRailways Statistic). TrafficDnsity: Passenger-kmper Route-kmnper Day

Tabl IV-20:JR EastJWn (1991) Traffic Density Rail Revenue (thousandpersons/km) (VBillion) Shinkansen 52.4 428 Conventionalline 45.1 1,290 Mctropolitan 187.1 935 Local 17.2 455 Total 45.9 1,718 Sour IjR.East apan

Ta IV-21:Transpolt Volume Honshu as PRs FY 1987 FY 1991 FY 1987 FY 1991 Passenger-km 191.4 231.8 104.4 112.7 (bAiLionperson-km) SourceMinistry of Transpor Tsudo Toki Nenpo(Anmual Rlways Statistics).

TableIV-22 CarFrequency Honshu IMsae PRs FY 1987 FY 1991 FY 1987 FY 1991 Car Frequency (thousandtimes) 249.7 297.6 553.2 619.9 Avragc Car per Train (car) 8.1 7.8 5.5 5.9 Soure MinistryofTransport. Tctsudo Tokei Nenpo (AnnualRaihlays Staisis). Car FrequencyqCarkan per Route-kn per Yenr

Tasb IV-23:Employee per Unit Honshu JRs lare PRs FY 1987 FY 1991 FY 1987 FY 1991 Rail Ernployeek 151 134 56 56 (thousandpersons) Employee per Route-kmn 102 9.2 19.6 19.5 (persons per kIn) Emnployeeper Car-km 40.9 31.0 35.5 31.4 (persons per n) **(1.15) (0.99) (1.00) (1.00) Employee per Passenger-km 0.79 0.58 0.54 0.50 (persons per million person-kin) *(1.46) (1.16) (1.00) (1.00) Sourc MinistryofTransporr. Tersudo Tokei Neupo (AnnualRailways Staistics). Notes lathisCapter.vecomparehenumberofemployes pernitrinordertoanalyzelaborproducrviy.TyismerhodisgeneallY used injapan. in Chaptr V. dt lbor producrivityof allJR compamisis comparedto lre PRs. whilein this dcptrta of the dee HonshuJRs OR East.JR Cenra and JRWesc)is analyzed. Rail Employee permanentemployer- not induded pa-t-ime employee Tke figuresin parenthesisis the raio to the figuresfor the large PRs Japanese National RailwaysPrivatizaton Sudy II 67 reachingonly 20% ofthe HonshuJRs. The reason higher dtan tiat for the HonshuJRs (HonshuJRs that the route-kmof the LargePRs is short relative 54 personsper kIn, LargePRs: 63 personsper km). to the HonshuJRs isthat whiletheprivate railways' (Comparisonover dte period fiom 1987 to 19911 maintrain linesare within the larg cities,the JRs' Trort Vo&me train lines are a combination of innercity routes, intercirytrunk routes and locallin routes. 4.55 ThenumberofpassengerstheHonshujRs 4.50 Thenumberofpassengerscarriedperyear carried has grown from 191 billion person-kmin byboththeHonshuJRsandlargePRscomesto8 1987 to 232 billion person-km in 1991, at an billion. However, in terms of passenger-kIm,the averagerate of 4.9 % over the 5 ycar period.This Large PRs carried 113 billion, not evcn half of growth rate is higher than that of the Large PRs HonshuJRs' figure (232 billion).Longer traveling (from 104 billionperson-kmto 113 billionperson- disance per passengerfor theJRs than th for the kIm,1.9% increaseper year). LargePRs contributed to this lifference. Improumuentin Sewce 4.51 As a sharc of total domestic transport services(in te;msof passnger-km), the railwaysa 4.56 hle Honshut JRs' esquencyof car opera- awholchold 30.1%, 17.4% ofwhich isheld by the ofon(car-km per route-n) islowconmpared to that Honshu JRs, and 8.5% by the Laurc PRs. In the of the Large PRs and the averagenus ber of cGaS three rnajormctopolitan areas,the railways'share connected to form one train tends to be high. In is considerablyhigher at 50.6%, rcflecting the response to dte increasein aansport volume over importanceof its finction as mcansof transporta- the 5 yearperiod, however,whilc the Honshu JRs tion for the residentsofthc largccities. .lc ac have decreasedthe averagenumber of carsper amin number of passengerstransported per day by the (from8.1 carsper rain in 1987 to 7.8 carsper tmin privaterailvays (induding smallprivate railways) in 1991), they have increasedtheir car frequency totals22 million peoplc, 21.5% of all transporta- (car-km per route-kn) by 19% (from 250 thou- ion in thelagecities. Intheselargecitics, theshare sand timesin 1987to 298 thousandrimesin1991). Of passengerstransported by private itys Ls In comparison,the responseof the Iarge PRswas greatertan hat of the JRs (17.5%). to increasenot only thecar frequency(by 120h)but 4.52 The avragc trip distanceof passengeris also the number ofcars per train (from5.5 carsper longerrEJRs (HonshuJRs:28km,LargePRr 14 train in 1987 to 5.9 cars per train in 1991) during In). This reflectsthe sitation decribed above this period. The increasein car frequencycan be wherethe lage PRsare concentratedwithin cities, seen as ainimprovement in pasenger suvicc, fur- whilethe JRs servethose travelingbetween cities, ther incasing tEUnsportdemand. making traveling distnce per passenger much Numbr .,pl,usper Unit longer. 4.53 The trafficdensity (the passegcr-kin per 4.57 The Honshu JRs reduced the number of routc-km per day) of the Hoashu JRs islow, only cmployeesin the raiload sectorduring this period, 40% of the figure of the LargePRs, orwingto the resulting in a reduction of 17 thousand workers &ct that theJRs have many local tain lineswhile overtheSyeaw (from 151thousandin 1987to 134 the Large PRs servemainly inner city commuters thousand in 1991). The number of employeesat (LargePRs: 107 thousandpersons per in, JRs: 44 the LargePRs was almost unchanged over the same thousand persons per km). If we divide JR East period. Japan's conventionallines into those witdin the 4.58 In terms of the number of employeesper Tokyo metropolitanarea and locallines, the met- route-kmn,the figure for the Honshu JRs is below ropolitan lineshave a trffic density of 187 thou- thatofthe LargePRs (HonshuJRs: 9.2, LargePRs: sandpersonsperkm,alevelsinilartothatofthe8 19.5). This reflecs the differeat nature of their Large PRs in the Tokyo area (153 thousand per- respectiveroutes; while the Honshu JRs operate sonsper km), while for the locallines the figure is innrcity lines, interciy lines and local lines, the verylowatl7thousandpersonsperkm.Becauseof Large PRs operate only innerciqylines. For ex- the locallines, the trafficdensity ofJR East Japan ampie, JRs' distnces becween stations (average is lower than tht of the Large PRs. distance between sations: Honshu JRs 4.3 kin, 4.54 Intermsoftheavergenumberofpassen- Large PRs 1.6 kIn) are longer than those of the geusper car-Ian,the figurcfor the LargePRs isalso Large PRs. 68 JapaneseNational RailwaysPrivatization Study II 4.59 Ifwelookatnumberofemployeespercar- is reflectedin the figures. In particular, the JRs kim,in 1987, immcdiatelyafir privatization,the operate many local lines, which, by nature, have figureof the Honshu JRs was 15% higher than lowtrafficdensityandthereforerlativelyhighunit that of thc Large PRs. However, if wc compare costs. In this contaxt,a dosc comparison of costs the two in 1991, the number had becomeroughly betweentheJRs and the LargePRs suffers from the equal (HonshuJRs: from 40.9 in 1987 to 31.0 in same difficultiesof the productivitycomparison. 1991, Large PRs: from 35.5 to 31.4). The JRs' initial inferiority to the PRs in terms ofemploy- ecs per car-km was a result of their low car 4.63 Interrnsoftheavcragecostoftransporting frequency, despite their low number ofemploy- one passengerover one kilometer, JRs' figure is ecs per route-km. In the 5 yearsfollowing JNR's higher. Comparing 1992 in order to avoid the privatization, the JRs were able to increasetheir influenceofrthe Shinkansen's lease fec, the Honshu car frequency, while decreasing the number of JRs requiredV112.5 per passcnger-kmcompared to cmployees.Thus the JRs were able to raise their Y8.9for the LargePRs. Breakingdown costs,we labor productivity, operating their cars with a can see that JRs' unit costs for personnd expenses, similarnumber ofemployces to that of the Large repair expensesand depreciationcosts werchigher PRs. (1991 figures).Therewaslitdedifferences inpower 4.60 Comparingthe employeesper passenger- expensesor the amount of ax paidL Iam,the LargePRs, which operatewithhigh traffic CostperCar-km density,high car frequency and high av4cagupass sengersper car, havea lower ratio. In this respect, 4.64 Intermsofcostpercar-kr,JRs'unitcost Lheirproductvityisgreaterthantharofthe Honshu is also higher than that of the LargePRs (Honshu JRs. However,over the 5 year period, the Honshu JRs: F725 thousand, Large PRs: Y541thousand). JRs were able to increasetheir transport volume However, breakingdown the aggregatecosts, we (passenger-kin)by 21%, at the sametimc reducing fmdtat JRs unit costsforpersonnel and power are theirwork forceby 1l%.Asaresult,the ratio ofLhe lower than those of the LargePRs. In the previous HonshuJRstotheLargePRs&Ilfroml.46in 1987 comparison made in terms of trnsportng one to 1.16 in 1991.Thisdecline represensa shrinking person overone ian, the unit cosCfor the JRs was of the gap in labor pmrductivitybetween the JRs higher dueto te' low number of avcragepassengers and the LargePRs. per car and low car frquency. However,compar- 4.61 In conclusion,analyzing number of em- ing costs per car-km, which serves to exdude the ployeesper unit by the simpleaverge comparison, diff&renceof avcragepassenger per car, there is a the productivity of the Honshu JRs improved reversem sone of the items. greadyin the fiveyears foilowingJNR's privatiza- ponnel CosapffEmployee tion and reachedon a parwith LargePRs in certain indicators.However, the gap betweenthe JRs and 4.65 IntheperiodimmediatelyfollowingJNR's the LargePRs still exists in ernployeeperpassenger- pnvatization,JRs'personnel costperemployeewas Ian, although this does not necessarilymean rhat lower than that of the Large PRs. However, by JNRislessefficient.Thisindicatorcould reflectdhe 1991 the figurehad risen to the almoscsame levcl two factors';(a) JRs operatelocal lines, too, and (b) of the LargePRs. The averageage of employceswas PRshavenotbeenaseagerasJNRfortheexpansion 3 years greater at the Honshu JRs (all scaors of transporation capacity, resulting in more includingsidebusinesses)inl991. Whiletheaver- crowdedcars at the expcnseof passengers.Later in agc agerose at thcJRs over the 5 year perioddue to ChapterV, we compareproducivity of the twoby thc 5 yearextensionofretiringage,at the LargePRs regressionanalysis which takesinto account of the it dedined. A number of JR's aged workerswill diflcrcnceof networkand service, begin to retire5years laterand restructuringvadl go on the way.

4.62 Below,we comparethe cost strucure of See'Dai ToshiKen Shitctsu no SezubiToshi ni Tsuite theant HonsujRs the lrge PRs In makng di(Invsmenmt of PrivateRailways in MetropolitanAreas).' theHonshuJRsandtheLargePRs.Inmaking this 1991.and 'TesudoUnchin, Shushi to SetsubiTos"i comparison,itis imporant to bearin miid thatthe (RailroadFams. Profiability and Capiml Invesument).' di&rence in operatingconditions between the two 199Z,by Hideli Moriya,the Japan DevelopmentBank ]a mmNational RailwaysPrnatization Study U 69 rableI-24: Costper Passenger-km

Honshu JRs Lae PRs FiscalYear 1987 1991 1992 1987 1991 1992 Operating Expenditure 14.2 13.5 12.5 7.5 8.5 8.9 PersonnelExpenss 4.2 4.3 4.5 3.6 3.8 4.0 Repair Expenses 1.6 2.5 2.2 0.9 0.9 1.0 PowerExpenses 0.6 O.6 0.6 0.6 0.6 0.6 Other Expenscs 5.5 3.7 2.2 0.9 1.2 1.2 (Leaseon Shinkansen)* (3.7) (1.6) (-) - - () Taxes 0.3 0.4 0.5 0.4 0.4 0.5 Depredation 1.9 2.1 2.5 1.3 1.6 1.7 Source:Ministry ofTmnsport. Tetsudo Tokci Nenpo (AnnualRailways Statistics). * lease on Shinknsen Expenseis fiom de financialstatemnts ofthe ShinkansenHolding Cooperation. rabl IV-25:Cost per Car-km 11'Thousand Honshu Rs Large PRs FiscalYear 1987 1991 1987 1991 Opeaing Expenditure 735 725 499 541 PersonnelExpenses 218 230 236 241 RepairExpnses 82 131 57 59 Power Expenses 33 32 38 36 Other Expenses 287 199 59 74 (LeIason Shinkansen)* (193) (83) - - Taes 14 21 25 28 Dcepration 101 111 83 103 Source MinisnyofTraspoff. Terudo Tokei Ncnpo (AnnualRailways Smtistcs). -eI on Shnkansen Expenseis fiom dte financidalsaenents ofthe Shinkan Holding Coopernaon.

TableIV-26 Persomel Cost and Age (MlUon,Yea) Honshu JRs ue PRs FiscalYar 1987 1991 1987 1991 AveragePersonnel Cost 5.3 7.4 6.7 7.7 AvemrAge' 37.8 41.2 40.0 38.3 Source Ministy ofTranspor. Tersudo Tokei Nenpo (AnnualRailways Srasics).JRs and PrivareRaihways. AveragePersonnel Cost - PersonnelCostslNumberofEmployces of RailroadSector Al Secors induding railroadand odterside businesses

TableIV-27: Fare Rate

Honsu JRs Lae PRs 1987 1991 1987 1991 Pa* 6.8 6.6 5.2 5.7 (1.31) (1.16) (1.00) (1.00) Non-Pass 21.1 22.0 14.7 16.3 (1.44) (1.35) (1.00) (1.00) Total 15.2 15.6 8.3 9.2 (1.83) (1.70) (1.00) (1.00) Sourie Mnisuy ofTmnsporr.Tetsudo Tokei Nenpo (AnnualRaiways Statistcs). * Pass Commutepass passenger Notes Fare Rare - PassengerRevenue I Passenger-kn.The figuresin parenthesisis the rtio to the figuresfor the LargePRs 70 Japanese National Raihlays Prvatization Study II FareRae the same period, inreasing fare pricesby roughly 4.66 In twrmsof fare rate, given by passenger 20%. Asa result,the differenc becweenthe fEs of revenuedividedbypassenger-kln,thefigureforthe the HonshuJRs and the LargePRs in urban areas HonshuJRs is higherat 1.7 timesthat of the Ire has been deacreasing. PRs(HonshuJRs:15.6, LagePRs V9.2).One of 4.68 AlongcertainlineswheretheJRscompete th reasons.hhih fuJR.rlativcl direcdywth the privateralways, they have intro. the reasonsbehind JRs relativelya hhigurs duceda specialfare discount system. Following the prernurn paid on excpresstravel and the Shnicase. tw *rcie mlmne ytelrePsoe In orderto achievea moresuitble comparison,we the fiveyear perod, wltheb theseccrtai lines the look at the fare ratc of commuterpass passengess. He the figureisV6.6 the HonshuJRsand V5.7 JRs presendyoffier cheaper travel in somecases. for the LargePRs, decreasingthe differencegiven FinanIPergmancc OperaingRewnie above to less than 1.2 times. Moreover, if we 4.69 TheratinvcRucofthCHonshuJRs comparethe figuresfor 1987 and 1991, wec grewat an averagerate of 5.5% annuallyover the that the gap has been shrinkingas a rcsult of the fiveyear period.This was due to the stronggrowth Larg fareapeotTasa.Rsincreases.roggr arge PR fireo p st.rivatiaon i o in the numberofpassengers (4.9% avrage annual 4.67 Prior to prmvanzation,in order semprove growth in terms of -assenger-kn).For the Lrge their revenueposition, JNR implementeda series PRs, the averager;. e of growth over the same ofprice hikes Asa result,their faresbecame cen- penod was 4.9% per year.This was due mainlyto sive as comparedto similar lines operated by the the efec of thc increascin re (fis incrcasedby privaterilwayswithizthemetropolanareas. The crease in fires caused passengersto slhifi a yearlyaverge of 3.4%). from JNR, necessitatingfuther fire increasesand Operaing Couts thus forming a vicious circde.Since privatzation 4.70 Thc sum of the Honshu JRs' operating (from 1987 to 1992), th JRs have not incrcased dnon-operatingcostsgwatanaageannual teir ofaes,apart from an incse for the iroduc- rate of 5% throughthe period. Ihc reasonsbehind hand,ftheooanrsumhw i se198 On twice m i relativelymodest growth were downsring in

TabeIV-28: FaU Rat (49 Between TicketType 1986 * 1987 199I 1992 Ueno One Way JRs 280 280 290 290 PRs 230 250 260 280 Ichikcwa Pas JRs 8,400 8,400 8,650 8,650 (onemonth) PRs 8,600 9,400 9,690 10,900 Shibuya One Way JRs 180 200 210 210 PRs 140 140 170 190 Kichijoji Pass JRs 5,400 6,000 6,180 6,180 (onemonth PRs 4,700 4,700 5,560 6,600 Shibuya One Way JRs 310 360 270 270 PRs 160 160 250 270 Yokohama Pass ]Rs 9,300 10,800 8,030 8,030 (onemonth) PRs 6,580 6,580 8,330 9,620 Nagoya Onc Way JRs 440 440 450 450 PRs 430 480 490 540 Yokkaichi Pass JRs 13200 13,200 13,600 13,600 (onemonth) PRs 11,500 13,270 13,570 15,800 Osaka One Way JRs 440 440 450 450 PRs 410 480 460 480 Nam Pass JRs 13,200 13,200 13,600 13,600 (onemonth) PRs 11,620 13,180 13,570 15,350 Sourot Insiini ofTwisponation Economics, Ssig DeMfm Te*udo(RaiFaoBok). a JRsfior 1986 areJNR JapaneseNational Ralways Privatization Study II 71 order to curb the increasein personncdcosts, a billionin fiscal 1987 to V26 billion in fiscal 1991 lighterdebt burden due to progress in the repay- (althoughin fiscal1992 it recoveredto V81 billion ment of debt inherited from JNR, and falling as a resultof the increasein fare prices). intcrestrates. Current profit grew from V170bil- lion infiscal1987 to V293billioninfiscal 1991.On11tio of C et Poit to Re the other hand, for the large PRs, the sum of 4.71 At 7.5%, JRs' ratio of current profit to operatingand non-operating costs grew at an aver- revenuein 1991was high relativeto the LargePRs age annualrate of 5.4% overthe same period.The (2.8%) as well as to the averageof all Japanese increasein depreciationcosts and interest pay- industry (3.0%). Its ratio fell in fiscal1991 due to mentsf611wingsuchlargescale investmentprojects lower profits caused by increasedcosts from the as quadrupletracking were the major faaors bc- purchaseof the Shinkansenline, but if we look at hind thisgrowth rate. Current profit fell from V42 the fiveyear trend it is apparent that the ratio has TableN-29; Ralway Profit and Loss of the HonshuJRB 11Bii0on,Percent) FiscalYear 1987 1988 1989 1990 1991 1992 OpeactingRcvenu^ 3,139 3,361 3,483 3,746 3,891 3,899 OperatingExpen4iture 2,710 2,865 3,004 3,210 3,137 2,927 OperatingProfit/Loss 429 496 478 536 754 972 (Ratioof Profitto Revenue) (13.7) (14.8) (13.7) (14.3) (19.4) (24.9) (Ratioof Profitbefore Depredationto Revenue) (25.5) (28.1) (24.6) (24.1) (31.8) (40.0) Non-OperatingRevenue/ -259 -259 -200 -155 -461 -705 Expenditurc* - CurrentProfidLoss 170 237 278 380 293 267 (Ratioof Profitto Revenue) (5.4) (7.1) (8.0) (10.2) (7.5) (6.8)

Table 30:Raily Profitand Loss of h LarpePRs (VBDiln,Percent) FiscalYear 1987 1988 1989 1990 1991 1992 OpentingRevenuet 921 988 1,012 1,053 1,110 1,196 OperatingExpenditure 786 828 869 918 963 1,003 OperatingProfit/Loss 135 159 142 135 147 193 (Ratioof Profitto Revenue) (14.7) (16.1) (14.1) (12.8) (13.6) (16.2) (Ratioof Profitbefore Deprciation to Revenue) (28.8) (30.4) (29.9) (28.9) (30.1) (32.1) Non-OperatingRevenue/ -94 -91 -84 -109 -116 -110 Expenditure Current Profit/Loss 42 69 59 26 32 83 (Ratioof Profitto Revenue) (4.5) (6.9) (5.8) (2.5) (2.8) (6.9) Source Ministry of Transporc. Tersudo Tokei Nenpo (Annual Railways Statistics). * OperatingRevenue: Fare revenucand otber evenue,not includedsubsidies Opting Expcnditurc Labor.energy, repair. miscellanus expenses.tax (not induded corporate tax) anddepreciation Non-OperatingRevenutE/xpenditurc is estimatedby multiplyingtotal Non-Operating Revenue/Expendikureby the mtio of the railwayfiwd assets o totl fied aets.

TableV-31: Business Proit tDTotl Catal (Percent) FiscalYear 1988 1989 1990 1991 1992

HonshuJRs 9.2 8.9 9.7 7.5 * 6.6 * Large PRs 5.1 4.8 5.0 5.0 4.4 Sourcc:* Ministryof Transport,Tersudo Tokci Nenpo (AnnualRilways Statistics).1992: JRs and PrivateRais Nooe=Fg aestotal businessmtsults iucludin& railway business and other side business. Total Capital = Current Assets + FuoedAssers + Deferred Assets+ Notes Discounted and Endorsed. Total Capital the average of the beginning and the end of fiscalyear. Business Profit - Operating Profit + Inteest and Dividends Received * BusinessProfirtoTotal Capitl ofhe Honshu]Rsdedi-nedin 1991and 1992 dueotw eincreaseofboth costand fiaxsed s by acquiringthe Shinkans line. 72 Japanese National Railwys PrivatizationStudy II been improving.While the unit costs oftheJRs ame 4.75 In the same period, theJRs werealso able higherthan thosc of the LargcPRs, they havebeen to downsizetheir work force by 11% and gready ableto remain higherprofitabiitywith theirhigher increaselabor produaivity. Due to the difference fare race. inthenature oftherailwaysopycmedbytheJRsand Reh*rxox As.ret the LargePRs, it is not possibleto nake a simnple comparisonof their performance.Howevcr, look- 4.72 In terms ofbusiness profit to total capital, ing at the average for the Large PRs and three which measuresthe return on assets,the Honshu Honshu JRs in 1991, in terms of car-km, the JRs JRs(1992:6.6%)outperformtheLargePRs(4A%). were able to decrase their number of employees Although the 1992 ratio for the Honshu JRs is per unit to a level simnilrto that of the LargePRs. influenced to somecxtent by the purchase of the In termsof passenger-kmn,while still below&hat of Shinkansen line, which EripledJRs' assets, this the Large PRs, the JRs were able to decreasethe prowvdeo detract litde from its profitability. difrence in number of employeesper unit fiom FncdPosidan 46% in 1987 to 16% by the end of the period. (In this chapter,simple average method isus5U for the 4.73 Vhile the JRs have been ablc to achieve compaison of the two, and in Chapter V produc- rdatively stable profitability,their financial posi- evity is compared by rgression analysiswhich ton weak In 1991, when theassets and liabilities takcsint account the nctwork condition and ofthe Shinkansenwere added onJRs balancesheet serviceoutput.) folowingJRs'acquisitionoftheShinkansenlincin 4.76 In terns of unit cost, JRs' costs are still dhat year, JRs ratio of net worth to total capital higherthan thatoftheLargepRs. Bycompens beame 7.3%. This isweil below hat ofthe private for their higr costs through setng higher fares, raiys (17.3%), though the ratio for thesecom- thcJRshave been able to mainain a higher profit- paniis is low relativeto other industies. abilitthantheLargePRs.However,overthe5year zssgcbBon period underconsideration, while theJRshave not raise didr fares,the price offaresfor theLarge PRs 4i.74 In the 5 yers fol ng priatization, the has risenbyrougbly20%.As aresult, thedifference Honshu JRs haveseen a 21% incrase in passenger in fiarepriccs betwe the two groups has been demand. While the strong economy during tbis deidiuug and, within some areas servedby both, period is behind this increase,JRs' own eiiorts can thc JRsare now carryingpassencrs at lower fares been seen to stimulatedemand. For example, the tan th of the private railways. increasein car frequencyachieved, while decreas- 4.77 In comparing the performance of post- ing the avrwagenumber of carsper train, shows an privatiationJRs and the large PRs,we can seethat, improvement in service, contributing to the m- in termsofservice,producti and fires,theJRs are creascin demand. approachingthe standard setby the Lagc PRs Japanese National RailwaysPrivatization Study 11 73

TableV32: Balance Sheet of theHonshu JRs (fBillion) FiscalYear 1987 1988 1989 1990 1991 1992 CurrentAssets 629 543 557 675 676 669 FixedAssets 5,433 5,286 5,334 5,446 14,636 14,566 TotalAssets 6,062 5,830 5,891 6,121 15,312 15,235 CurrentLiabilities 1,183 1,254 1,241 1,281 1,494 1,403 FixedLiabilitics 4,216 3,831 3,756 3,805 12,695 12,629 Total Liabilities 5,399 5,085 4,997 5,086 14,189 14,032 CapitalStock 412 412 412 412 412 412 Lcal Reserve 205 205 205 205 211 215 RetainedEamings 46 127 277 418 499 576 Shareholder'sEquity 663 744 894 1,035 1,123 1,203 Ratio:Net Worthto TotalCapita] 10.9% 12.8% 15.2% 16.9% 7.3% 7.9%

TableIV-37: Balance Sheet of the LargePRs (I Billion) FiscalYear 1987 1988 1989 1990 1991 1992 CurientAssets 1,848 2,057 2,016 2,180 2,225 2,209 FixedAsset 4,504 4,870 5,375 5,918 6,467 7,101 DeferredAssets 2 2 1 0 0 0 TotlAssets 6,354 6,929 7,392 8,098 8,692 9,311 CurrcntLiabilities 2,167 2,216 2,266 2,440 2,606 2,727 FixedLiabilities 3,287 3,458 3,694 4,132 4,517 4,963 Rcserves 0 15 34 48 64 79 TotalLiabilities 5,454 5,689 5,993 6,619 7,187 7,769 CapitalStodc 478 640 710 742 745 753 LegalReseve 338 506 583 620 630 643 Rtained Earnings 83 94 106 117 130 146 Sharcholder'sEquity 899 1,240 1,399 1,479 1,505 1,541 Ratio:Net Worthto TotalCapitl 14.2% 17.9% 18.9% 18.3% 17.3% 16.6% Source MinisnyofTramspDfrTetsudoTokei Nenpo (Annu2lRilW2ya SSaitisrics). No=: Ratioof Net Worth to Toal Capital - Sharcholdes EquityI (Toal Liabilities+ Share.olders Equity)

TabkIV-34. Raio of NetWorh to TotalCapital (1991) (Pernt) HonshuJRs LargePRs AllIndustries AllManu&ctufi 7.3 17.3 28.8 38.2 Source:Ministyof Transport, TesaudoTokeiNenpo (inm aiRailwystaeics). HandbookofIndusvidFmnandaDara 1993by the JapanDevelopment Bank V. EconomicAnalysis of Productivity andPerformance of Privatized JapaneseRailways

ProductivityComparisonbetweenJRsand Private 5.3 Ther aretwoanaliamehodsforstudy- Railways ingproductivityin this research:(1) simpleavceage Analytical Metha& ofd Awdviy Comparison oDmparisonand (2) regressionanalysis. The simple betmm .P...... fdaxwaw. averagccomparison is the most common way to mcasureproductiv, in whih passenger-lanper 5.1 Onc of thc peculiar fiatrs of the Japa- employceand car-kanper enployec, and train-km .Aeserailways system is the coexstenceofprivarizd per emploe are often used. Since thismethod railwaysandpubLicralways.Thereareseveraltypcs does not takc into accountnetwodr characteristics, of owncrship,from municipal-ownedsubways or itprovidesonlyageneralindicationofproducivity the semi-publiclocal railwaysto purdy privare growth.The advantageofregression analysis is that owned railways.Against this bacground, an im- productvity measureentn reflects explicidy the pormantquestion is whether or not the JRs have diflrnces in nectwork,frequency of service,load been able to inprove their productivityfollowing factor, and ownership. privatizonwhencomparcdwith privaterailways, since productvity of public raihvaysis said to be X''P""u Growh lowerthat forprivat railways.There isacommon 5.4 There is no doubt that, following impressionthatproductivity inJRsinfacthasbecn privatisationof the Japan Raiway companieshave enhancedby a combination of privaization and been inceasing productivty remarkably.Hcre we deregulation. To hdp understand what facors willexaminehowproducaiviryofJRshasimproved mightcontribure to makeJapanese-styleprivatiza- and whether or not their productivityhas reached tion distinctfrom that of other reforms,it is essen- thc levelof Japaneselargc private railways,which tial to analyzchow much JRs have been able to are consideredthe most effcientorganizations in improveproductivity. The followingsections ad- the railwayindustry. dressthe questionwhat changesin productivityof 5.5 TablcV-1 showsthatlaborproductivityof JRs haveemerged following privatization ofJNR. JRs after privatizationhas increaseddrmadcally 5.2 Usuallythere are three difflreeninputs to undcr any kind of measurement:passener-km, assessproducivity growth ofJRs,such as (1) labor, car-km, and train-km per cmployee. Generally (2) energy, (3) roiling stodc Since railways are speaking, JRs' producaivityhas improved by a characterizedas a labor intensiveindustry, we will fictor of around chreetimes from 1981 to 1991. focuson labor produciviy, as measurcdby servicc For example, the producivity of the Japan Na- output per onc employee.Although changes in tonal Railwaysin 1981 was only 509 thousand labor quality and capital as well as technology passenger-lknper employce,which was less than might affact labor productivity, they are not one third that for largcprivate railways, but it had taken into account becausc data is not readily incasedwo 1558thousandpassenger-kmbyl991. availabe. Compared with the producmvitygrowth of largc Japanese National RailwaysPrivatization Study II 75 privatc railwaysfrom 1981 to 1991, which in- fore privatization,organizational reforms, espe- creased by less than 20%, productivitygrowth of cially labor rationalization,were implemented. JRs is surprisinglyhigh. Certainly privatization Relativelyspeaking, JRs after privatizationwere couldbeoneimportantdelmentinJRs' remarkable free to pursue their own developmentas the largc productivitygrowth. private railwaysdo and they are endeavoringto 5.6 The productivitygrowth of JRs was not increas produaivity . even and has varied from one period to another. 5.7 Althoughthe productivityof JRs has in- Furthermore, produaivity of JRs started to im- creasedsharply since the mid-80s,the productivity provesvceral years before privatization. In fact, the levelis still lowerthan that of largeprivate railways. productivitygrowth rate beforeprivatization was In fiact,the productivityofJRs is about 23% lower larger than that afterward. For example,the JRs' in passenger-km,6% lower in car-km, and 22% growth rate of productivityduring the frve years lower in train-kmnper employee less than thosc from 1985 to 1989, which indudes years before forlargeprivaterailways.However, data inTablc and after privatization,was 79.3% in passcnger- V-1 do not pay attention to any difference in kIn, 77.1% in car-kin, and 91.1% in train-lan per network conditions between JRs and large pri- cmployee.However, in the post-privatizationpC- vate railways. riod(from 1987to 1991)thesefigureswere37.4%, 34.5%,and37.9% respecively.Itprovedthathigh e m productivitygrowth occurred severalyears before 5.8 To evaluatewhat typesof activityare re- privatization.The reduction of employecshad al- sponsible for productvity increases,rail service ready begun in the late 1970sand the number of aciavitiesare divided into fivc categories:1) rail eMployeeswas reduced at about 10% everyyear. opeation, 2) station, 3) track maintenance,4) car This high productivitygowth could be attnbuted m aintenandceand 5) administration and engineer- to theffect on employcesiniNRofthe announce- ing at headquarters.The questons are; whether muentthatJapan Natonal Railwayswouldbepriva- privatizationcould irnproveproductivity in each tized sooner or later. The anti-privatizationside aaivity equalb, and, if not, whichof activity im- may have had provide an incentive to improve proved the most andwhi is the least, andalso in productivityinorder to showdictprivaization was what activitiesJR's have caught up with large not necearywhile for the pro-privatation side,it privateraihways. Herewe definelabor productivity might have provided an opportunity t promote by actvity diffience as serviceoutput per em- the rationalizationpln Duing the fw yearsbe- pIOYeCin eCh activity.

Tab V-i:Labor Productvity Comparison btwn JRand Lwt Pra LaborPruvty* by Actv;ty Passenger-kmper Car-km per Train-km per employee (thousand) employee(thousand) employee(thousand) Year JR LargePrivate JR La Private JR larwe Priate 1981 509 (0.30) 1693 (1.00) 11.7 (0.45) 25.8 (1.00) 1.36 (0.27) 5.05 (1.00) 1982 524 (0.31) 1710 (1.00) 12.0 (0.46) 26.1 (1.00) 1.41 (0.28) 5.07 (1.00) 1983 573 (0.33) 1738 (1.00) 12.8 (0.48) 26.5 (1.00) 1.52 (0.30) 5.08 (1.00) 1984 633 (0.36) 1751 (1.00) 13.6 (0.50) 27.2 (1.00) 1.68 (0.33) 5.13 (1.00) 1985 758 (0.42) 1793 (1.00) 15.7 (0.57) 27.6 (1.00) 2.02 (0.39) 5.15 (1.00) 1986 929 (0.51) 1837 (1.00) 19.4 (0.70) 28.1 (1.00) 2.56 (0.50) 5.15 (1.00) 1987 1134 (0.61) 1863 (1.00) 22.3 (0.79) 28.2 (1.00) 3.06 (0.60) 5.12 (1.00) 1988 1281 (0.67) 1922 (1.00) 25.0 (0.89) 29.3 (1.00) 3.55 (0.68) 5.25 (1.00) 1989 1359 (0.70) 1939 (1.00) 27.8 (0.92) 302 (1.00) 3.86 (0.73) 5.30 (1.00) 1990 1496 (0.75) 1985 (1.00) 29.6 (0.98) 31.1 (1.00) 4.11 (0.79) 5.19 (1.00) 1991 1558 (0.77) 2019 (1.00) 30.0 (0.94) 31.9 (1.00) 4.22 (0.78) 5.41 (1.00) %change (1981-85) 48.9% 5.9% 34.2% 7.0% 48.5% 2.0% %change (1985-89) 79.3% 8.1% 77.1% 9.4% 91.1% 2.9% % dage (1987-91) 37.4% 8.4% 34.5% 13.1% 37.9% 5.7% 76 Japanese National RailwaysPrivatization Study II

Tabl V-2:Proporlonof Employeesby ActivityDifference EMPo EMPt EMPr EMPh (Opertor & EMPs (Trackmain- (Car main- (Administration Conductor) (Station) tenance) maintenance) Engineern Year JR LargePrivate JR LargePrivate JR Lrge PrivatcJR Lage Private JR LargePrivate 1981 72.6% 25.1% 41.9% 34.0% 14.6% 14.6% 18.4% 11.6% 12.5% 14.8% (0.50) (1.00) (1.23) (1.00) (1.00) (1.00) (1.59) (1.00) (0.84) (1.00) 1982 12-A% 25.9% 41.6% 33.5% 14.8% 14.5% 18.6% 11.3% 12.6% 14.8% (0.48) (1.00) (1.24) (1.00) (1.02) (1.00) (1.65) (1.00) (0.85) (1.00) l9b3 12.9% 25.4% 41.1% 33.8% 15.0% 14.3% 18.1% 11.3% 12.8% 15.2% (0.51) (1.00) (1.22) (1.00) (1.05) (1.00) (1.60) (1.00) (0.84) (1.00) 1984 13.9% 25.7% 40.6% 33.4% 15.3% 14.4% 17.2% 11.2% 13.0% 15.3% (0.54) (1.00) (1.22) (1.00) (1.06) (1.00) (1.54) (1.00) (0.85) (1.00) 1985 -16.2% 25.6% 38.3% 33.6% 15.8% 14.5% 15.1% 11.2% 14.6% 15.3% (0.63) (1.00) (1.14) (1.00) (1.09) (1.00) (1.35) (1.00) (0.95) (1.00) 1986 182% 26.0% 36.9% 33.5% 18.2% 14.4% 11.8% 11.2% 14.9% 14.9% (0.70) (1.00) (1.10) (1.00) (1.26) (1.00) (1.05) (1.00) (1.00) (1.00) 1987 20.1% 25.9% 35.5% 33.6% 20.6°h 14.2% 8.6% 10.8% 15.2% 15.6% (0.78) (1.00) (1.06) (1.00) (1.45) (1.00) (0.80) (1.00) (0.97) (1.00) 1988 203% 26.2% 34.0% 33.7% 20.0% 14.2% 8.8% 10.5% 16.8% 15.4% (0.77) (1.00) (1.01) (1.00) (1.41) (1.00) (0.84) (1.00) (1.09) (1.00) 1989 20.7% 26.6% 33.3% 33.7% 19.5% 14.1% 8.8% 10.2% 17.8% 153% (0.78) (1.00) (0.99) (1.00) (1.38) (1.00) (0.86) (1.00) (1.16) (1.00) 1990 21.0% 26.5% 32.7% 34.0% 18.9% 13.8% 9.0% 10.3% 18.3% 15.4% (0.79) (1.00) (0.96) (1.00) (1.37) (1.00) (0.87) (1.00) (1.19) (1.00) 1991 21.3% 26.9% 32.8% 34.2% 17.9°h 13.7% 8.8% 10.1% 19.2% 15.1% (0.79) (1.00) (0.96) (1.00) (1.31) (1.00) (0.87) (1.00) (1.24) (1.00) % chwr (81-85) 10.2% 2.0% -8.6% -1.2% 8.2% -0.7%A-17.9% -3A% 16.8% 3.4% % change(85-89) 27.8% 3.9% -13.1% 0.3% 23.4% -2.8% -41.7% -8.9% 21.9% 0.0% % change(87-91) 6.0% 3.9 % -7.6% 1.8% -13.1% -3.5% 2.3% -6.5% 8.3% -3.2 %

Rgw V-1:Comparison of ChangeIn Acfvlty-bass Employees

private(1g991)

8 EMPh JR (after) _ EMPr * EMPt JR (during) _0 EMPs * EMPo JR (befoe)

0 10 20 30 40 50 60 70 80 90 100

Norm:JR (before):beiore privarizauion (1981) JR (during):during pricatizaion procs (19S5) JR (afir). akrerprivarzation (1991) Private(1991): privae railways (1991) EMPo:operator 8c conductor EMPs:statwn employee EMPt:track ainenanceeployee EMPr:-carmaintenance employee EMPhtadministrtion 8c engineering amployee JapaneseNational RailwaysPrivatization Study II 77 5.9 Bel;re analyzingproductivity by activity, trast, the proportion of administrationand engi- we willstudy changes in the proportionof employ- neering is dose to the level for large private ees by activity.Table V-2 shows change in the railways.These results may indicate that station structureofeach activity'semployees from 1981 to and car maintenance activities before privatiza- 1991.Stationemployeesaccountedforabout42% tion were inefficient and that JNR used many oftotal employmentin 1981,butin 1991this ratio more employeesthan the private railways.JRs wasreduced to 33%. Car maintenanceemployees, after privatization stil have more employeesin a categorywhich accounted fbr 18% in 1981 also track maintenance and administration activities shranktoonehalfby1991. These two aaivities in than do large private railways.This may suggest terms of employment decreased sharply in the that sinceJRs' network is far [onger than that of periods beforeand after privatization.However, the private railways,JRs need more track main- the other activitiessuch as operator and conduc- tenance.And alsodue to increasesin the number tor, track maintenance, administration and en- of headquarters after regional break-up, JRs gineering increased their relative shares in might not be able to reduce employment in employment during the elevenyears from 1981 administrativesections as they have done in other to 1991. When compared with large private activities. Figure V-I shows die changes in the railwaysin terms of proportion of operators and compositionof employeesin JRs befire privariza- conductors, that for JRs were about 80% of the tion, during privatzaton, and after privanzation levelof large private railways in 1991. In con- compare with private railways.

TabeV4- Labor ProducfWIy CmnWpson by Actii Dfffew Imhouwndsofpaswsnge km par emply) PRDo PRDt PRDr PRDh (Opeator & PRDs (Track main- (Car main- (Administration conductor) (Station) tenance) tenance) & eng6werirq Year JR Lre Privatc JR LupPnargePvatJR LIg Priate argePrvate JR LargePrime 1981 4043 6750 1214 4981 3486 11627 2761 14640 4079 11445 (0.60) (1.00) (024) (1.00) (030) (1.00) (0.19) (1.00) (036) (1.00) 1982 4213 6603 1261 5104 3537 11813 2822 15132 4175 11548 (0.64) (1.00) (025) (1.00) (0.30) (1.00) (0.19) (1.00) (0.36) (1.00) 1983 4434 6828 1392 5145 3819 12116 3163 15410 4461 11460 (0.65) (1.00) (027) (1.00) (032) (1.00) (0.21) (1.00) (0.39) (1.00) 1984 4545 6806 1560 5243 4141 12152 3682 15679 4869 11446 (0.67) (1.00) (0.30) (1.00) (034) (1.00) (0.23) (1.00) (O.43) (1.00) 1985 4689 7016 1979 5343 4800 12395 5016 16058 5203 11750 (0.67) (1.00) (037) (1.00) (039) (1.00) (031) (1.00) (0.44) (1.00) 1986 5116 7063 2516 5478 5103 12738 7838 16453 6233 12356 (0.72) (1.00) (0.46) (1.00) (0.40) (1.00) (0.48) (1.00) (0.50) (1.00) 1987 5302 7202 3008 5540 5182 13168 12442 17328 7014 11954 (0.74) (1.00) (0.54) (1.00) (039) (1.00) (0.72) (1.00) (0.59) (1.00) 1988 5916 7335 3531 5703 6000 13527 13712 18389 7134 12471 (0.81) (1.00) (0.62) (1.00) (0.44) (1.00) (0.75) (1.00) (0.57) (1.00) 1989 6179 7303 3837 5748 6537 13710 14522 18933 7178 12657 (0.85) (1.00) (0.67) (1.00) (0.48) (1.00) (0.77) (1.00) (0.57) (1.00) 1990 6678 7481 4291 5832 7445 14415 15572 19336 7664 12918 (0.89) (1.00) (0.74) (1.00) (0.52) (1.00) (0.81) (1.00) (0.59) (1.00) 1991 6864 7504 4465 5899 8153 14727 16644 20044 7622 13394 (0.91) (1.00) (0.76) (1.00) (0.55) (1.00) (0.83) (1.00) (0.57) (1.00) %clnge (81-85) 16.0% 3.9 % 63.0% 73% 37.7% 6.6 % 81.7% 9.7% 27.6 % 2.7% %cdange (85-89) 31.8% 4.1% 93.9% 7.6% 362 % 10.6% 189.5% 17.9% 38.0% 7.7% %change (87-91) 29.5% 4.2% 48.4% 6.5% 5733% 11.8% 33.8% 15.7% 8.7% 12.0% 78 Japanese National RailwaysPrivatization Study 11 5.10 The analysisindicates that JRs improved than 60% of the levelof large private railwaysin productivityin stationand car maintenanceactivi- 1991.Thisshowsasharpcontrastwithcheprevious ties, but could not do so in track maintenanceand studyofMizumani, in which therewas not a signifi- administradon activities.Table V-3 supports this cant diffience between private and public rail- surmise.The inteesting point is that there wasnot ways. JRs' large scale network, the operation of much differencein productivityper operator and Shinkmnsen,andregionaldivisionofmarketsmight conduaor becweenJRsand largeprivate railways in requiremore administrative employees. Moreover, 1991.This ist-ue of the periodswel beforepriva- the competition in technological development tization. For example,in 1981 the diffirence in among regionalJRs might need to maintain more operator and conductor'sproductivitybecveenJRs engineers at headquarters than private railvays. and largeprivate railways was small compared with However,a 43% differencein productivityis still differencesin other activities.This result is very large cnough to justifylooking for other reasons. similarto the previouscomparative study ofJapa- "nseurban private and public radlwaysby MPzurang ,-I11. 5.13 Here we will analyzeproductivity difier- . .1 The productivitydiffirence in mainte- enceamong six JRs. To evaluatcthe privatization nanceactivities, especially track maintenancr,is effectonJRs' productivi, it iSintcrestng to study scilhlargebecwecnJRsandprivaterailwvays.Produc-the difference betwecn the three largestJRs in tivity per track maintenanceemployee in JRs in- Honsyu (henceforthreferred to as 'Honsyu JRs") creasedmore than two times comparedwith the andthe three-islandsJRs(hceforth referredto as Levclbeforc privatization in 1981.However, it is 3-islandsJRs"). Iabor produccivityof larkepri- stillonly 5 5% that for largcprivae railway The vate railwayscould bc used as a bendunark to possiblereason why JRs' productivityin tack compareproductivityamongJRs.TableV-4shows maintenanceremains lower than forlarge private laborproducxivityamongJRintermrofpasseager- rilways might bc that JRs have lrger networks kamper employcc. (e.g longerline haul and morelina) dan large 5.14 As shownin TablcV-4, there is a large privatcrailways. Furthermor the contracting-out diffirenc in labor produaivity between Honsyu schenein maintenanceactvities, whih largepri- JRsand 3-islandsJRs. In 1991,labor productvity vate raihlaysadopt, mightbe anotherreason for ofHonsyuJRswasabout2.8 times hihrtan that privateraiwayis' hih productivity of 3-ilands JRs.Honsyu JRs' prductivity is ap- 5.12 JRs' productvityper admini and proachingthat for large private railways.The aver- engineringemployee at headquartersis alsopoor age productimtyfor HonsyuJRs in 1991, for comparedwith privaterailways, which was less example,was about 86% that of largc private

TS*leV- LabWorProductity Compafson AmO JRsAft Pvafon (Thousandsof paspger-kmsper emnoyee) Honshu3 JRs 3-IslandsJRs PrivateRaiway TR TR JR JR TR TR jaR Smal Year East Central West Avg. HJcaido Shkkn Kyusyu Avg Privae P te 1987 1293 2034 918 1268 324 444 562 449 1863 580 (0.69) (1.09) (0.49) (0.68) (0.17) (0.24) (0.30) (0.24) (1.00) (0.31) 1988 1511 2245 1003 1442 375 595 584 499 1922 594 (0.79) (1.17) (0.52) (0.75) (0.20) (031) (0.30) (0.26) (1.00) (0.31) 1989 1636 2329 1014 1518 377 600 683 539 1939 593 (0.84) (1.20) (0.52) (0.78) (0.19) (031) (0.35) (0.28) (1.00) (0.31) 1990 1725 2608 1125 1647 428 653 740 592 1985 482 (0.87) (1.31) (0.57) (0.83) (0.22) (0.33) (0.37) (0.30) (1.00) (0.24) 1991 1868 2545 1162 1728 442 679 796 624 2019 498 (0.93) (1.26) (0.58) (0.86) (0.22) (0.34) (0.39) (0.31) (1.00) (0.25) %change (87-91) 44.5% 25.1% 26.6% 36.3% 36.4% 52.9% 41.6% 39.0% 8.3% 14.1% Japanese National RailwaysPrivatization Study II 79 railways.On theotherhand, 3-islandsJRs'produc- than Honshu JRs in the fiveyars since privaiiza- tivity on averageis close to that of less-effident tion in 1987. When comparedvwth smallprivate smallprivate raiways. This may reflectthe differ- railwayswhich have very similardemand condi- ence in size betwcenJRs and the private railways. tions, the growth ratesand levelof produaivity of Even JR Shilokt, the smallestof the JRs, which 3-islandJRs are extrememyhigh. Although there is employed46 thousand workers in 1991, is larger stil alarg differenceinlaborproductiviy between than Kintetsu, the largest private railwaywhich Honshu and 3-islandjRs, privatizationcould spur employed 11 thousandemployees in 1991. Thus, productivitygrowth cven in 3-islandsJRs, which 3-islandsJRs may still have surplus employeesin face dficlties stemming from the smaullsizc of relation to their size of sevice output. marketsand competitionfrom the road secEor. 5.15 Privatizationwith regionaldivision seems to havea verypositiveimpact on the enancemenbt of the productivity of 3-islands JRs. As far as 5.16 To clarifywhat factors were responsible prductivity growth is concened, the growth rate for the improvementin labor productivityin JRs, of 3-islandsJRs has seen, on average,3% higher regressionanalysis is appliedher TableV-5 shows

TbIS V.5:Labor Prodkctiviy Using ProduWvity ode(Thousnd PaRnger nEn #w) Owner- PRDo(operator PRDs PRDt(uac PRDr(car PDh(admin. & Metod ship &conduor) (station) mainuance) min_tnance) engineering) Case-i: JR 6864 4465 8153 16644 7622 (0.91) (0.76) (0.55) (0.83) (0.57) Simple avage Private 7504 5899 14727 20044 13394 (1.00) (1.00) (1.00) (1.00) (1.00) Case-2: JR 7434 4740 8380 19513 7944 (1t38) (0.78) (0.84) (1.21) (038) Mizutani (1993) Private 5393 6105 9985 16173 20877 (1.00) (1.00) (1O00) (1.00) (1.00) Case-3: JR 6970 4249 7615 16138 8411 (1.20) (0.79) (0.65) (0.69) (0.86) This study, Private 5793 5353 11788 23316 9762 (1.00) (1.00) (1.00) (1.00) (1.00) [Note] (1) (A) Case-I is rsults without controllingnetwork conditions:Result are ave for JR(6 pasc3ger JR raways) and for Prvae(15 a6rgeprive railways)in 1991. These ar the same shownin Table V-3. (B)Cae-2 is resultswithcontrolling nctwodr conditions butusingaprivuticproductivitycomparon model:Results for JR are sampleaverage and resultsfir privateare obtainedby substinstingJR'snetwork vaiables in Mizuranimodel. (C) Cas-3 is the resus ofa private-JRpwductvity compaison modd when networkcondiions wereconolled: Resultsof JR and Privt areobtained by assigningthe dummyvatable O) aone in the caseofJRs and zero ifJRs wee assumedto be opented by a pive compay. (2) (A) Equationsfor productivityesimation in Case-2ate as follow Mfr.uauuirshu' (1993) PrIsv&&-PuhL&Conzpwirx Me (a) ln(PRDo) = 5.725 + 0.478*1n(Qpkm)- 0.3781n(N) - 0.284*1n(r)+ 0.196In(SI) - 0.164'PUB (b) Wn(PRDs)- 5.704 + 0.576*1n(Qpkmn)- 0.658In(Ns) - 0.127ln(Rus) - 0.41rPUB (c) In(PRDt) - 5.089 + 0.433*1n(Qpkm) - 0.364,n1nr) - 0.088*ln(Ru) + 0.864 1n(CNT1) - 0.1600PUB (d) In(PRDr)- 5.629 + 0.423'1n(Qpkn) - 0.5581n(NI) + 0.636Pln(CNT2) - 0.405*PUB (d) Iu(PRDh)- 5.686 + 0.5021n(Qpkm) - 0.028Wln(Nl + 0.272'ln(lr) - 0307*ln'(Ns) - 0.104ln(Ru)- 0.118PUB (B) Equationsforproductivityesimation in Case-3areas foiows: Thi s,: Prih*-JR Compi*ox Modd (a) ln(PRDo) - 5.194 + 0.512'1n(Qpkm) - 0.4411n(N) - 0.294"1n(1r)+ 0205"n(SI) + 0.185'JR (b) ln(PRDs)- 7.719 + 0.443'In(Qpkm) - 0.463'In(Ns) -0.0171n(Rus) - 0.231'JR (c) ln(PRD) - 5.446 + 0.4#iln(Qpkm) - 0.361'ln(Lr) + 0.004¶Pn(Ru)+ 0.743(CNTI) - 0.437JR (d)In(PRDr) - 5.238 + 0.418'1n(Qpkm) - 0.427'ln(NI) + 0.74Vln(CNT2) - 0O368R (d) In(PRDh)- 5.523 + 0.555'ln(Qpbm - 0.4861n(NI)- 0.496'ln(Or) +0.16201e(Ns) - 0.034*ln(Ru)-0.149'JR whereQpkm - passenger-km NI - number ofline Ir - route-kn per line Si - load ictor(%) Ns - nunber of sation Rus - % ofunesround sation Ir oute-km R - % of underground CNTI - contacting-out fortadc inea CNT2 - contraaing-out for cr mantnac PUB - ownershipdummy(public-1) JR -JRdummy R-1) 80 Japanese National RailwaysPrivatization Study II the resultsof activity-basedproductivity compari- 5.18 The first row (R) of Case 3 shows labor son. Asproductivity measurements, labor produc- productivityofJRswhichis obtainedby regression tivityper operator and conductor, per station, per analysisusing data fromsixjRsandprivare railways track maintenance,per car maintenance,per ad- on the number of employees,passenger km, num- ministration and engineeringwere used. In this ber of lines, route km and others in 1991. The analysisdifferences in networkcharacteristics and secondrow (Privatc) of Case3 is alsothe estimated ownershipare consideredexplicidty Passenger-km productivityofJRs obtained from the simulation (Qpkm) as serviceoutput, load factor (SI)as service model basedon the sameidea as the secondrow of quality,number of line (NI), route-kmper line (Ir), Case2. The resultsshow the hypotheticalproduc- numberofstation (Ns),percentage share ofunder- tivity of JRs when it is assumed that JRs were ground station (Rus) as nenworkcharacteristics, operated by a private companywhile they main- contracting-out for track maintenance (CNTI), tainedthesamelevelofserviceoutputandnetwork. contracting-out for car maintenance (CNT2) as The esults indicate how productivity might be company policy characteristics,and ownership affected by the differencein ownership. Most dummy (PUB),and JR dummy JR) as ownership resultssuggest thatJRs'producivity isstil 20-30% differenceare taken into accounthere. less tan tat ofprivate ilways,except productiv- 5.17 Table V-5 comparesthe resultswhich are ity ofoperatorsand conductors.This suggeststhat estmatedby three differentmethods basedon the JRs stil rin more workrs than privatermiways cross-sectiondata in 1991: The first and second in most activities,evcn houghnetwork and other rows(JRandPrivatc) ofCase I compareproductiv- conditionsarc explicidycontroLld in the estima- ity by the simple avera which was obtained by tion. Specificaly, productivity of trak mainte- dividingtotal passenger-kmnby tocalemployes on nanceand car maintenanceis respectively35% and an activity-basis.The first row mR)of Case 2 is 31% lower than those for private vailwys.In labor productivity of JRs which was estimated conrast withsimpleaverageproductivityanalysis, usingMizutanisprivare-publiccomparisonmodel. productivityfor administrationand engineeringis Th coeffidens in the modelwere obtained by the only 14% less than that of privaterailways, when cross-sction data of public and private railways. diflerencesin netwodrkconditions are allowedfor. To estimatelabor productivityof JRs, the average Unfortunatelythe t-staticsfor each coefficientare passenger-kin,average number of employees,and about one owing to the small sample and it is the averge of network chaacristics among six diffialt to specifywhatelements contibute to the JRs were used.The secondrow (Private)of Case2 low productivityof JRs. But it may be safe to say shows the resultsof simulationwhen it is hypoth- that becausc of large-salenetworks JRs have tO esized that JRs' productivitymay be improved if retain a larger number of employeesin mainte- JRs wercoperated by a privatccompany. They arc nance and stations actinities,which may lower obtained when the dummy variable (PUB) is as- labor productivityof JRs comparedwith private signedzero in the simulation.In other words, the raiys. simulation indicates how ownership difference 5.19 Table V-6 is the results of simulation might affict JRs' productivityifJRs were to main- showingwhatextentHonshujRsand3-islandsJRs tain the currnt serviceoutput and network and may have redundant employees.Thre diffi&rent operateunder a hypotheticalprivate company scenariosare introduce& (1) pCrfct privatization,

Tabf V-6:Compadso of Emploes Among JRs: After Ptlution case Honshu JRs 3-islandsJRs policy option before afier changc beore after chge Senario 1 (perict privasization) 45,188 38,603 -6,585(-14.6%) 7,784 6,738 -1,046(-13.4%) Sario 2 (decse lines by 10%h) 45,188 44,566 - 622(-1.4%) 7,784 7,677 -107(- 1.4%) Sceario 3 (decse line h1u by, 10%) 45,188 44,252 -936(-2.1%) 7,784 7.599 -185 (-2.4%) Scenario4 (inese cona out by 10%) 45,188 44,755 -433(-1.0%) 7,784 7,662 -122(-1.6%) Japanese National RailwaysPrivatization Study II 81 (2) abandonmcnt of lines, (3) reduction in line in major citicssuch as Tokyo, Osaka, and Nagoya, haul, and (4) extensiveuse of contractingout for JRs in Honshu retain natural monopolisticpower maintenanceworks. The estimated productivity in intercity passengcrtransportation markets as under perfectprivatization scenario is obtained by well as in commuting markets in metropolitan assumingthat JRswere operatedas a privatecom- regions.However,JRs facefiercecompetition from pany when JRs maintained the current level of road ransportin citiesand fromair transportin the scrviceoutput and network In thiscase. the dummy long distance transportation narkets. Their mo- variablc OR) of ownership is assignedto zero in nopolisticpower has been waning rapidly in long Private-JRComparison Model. According to Sce- distancemarkets. nario 1, Honshu JRs as well as 3-islandsJRs could 5.21 Secondly,privatization has involvcda re- reduce employeesby about 15% of the present gional break-up into six passengerJRs and one numbcr on averagcif they were operated by the nationwidefreightJR. The regionaldivisionofJRs hypotheticalprivate company. As shown in Sce- has now been tumed to advantage, since it has nario 2, even if JRs abandon somclines, it would enhanced indirect competition among JRs and nor greadycontribute to improvingproductivity. direct competition between JRs and private rail- The declinein employment is only 1.4% in both ways.Since there areanumber ofparalel commut- Honshu JRs and 3-island JRs, when we assumed inglines in major dtieswhere both privaterailways thatJ Rs would reduce 10% of the number of lines andJRs providecompetitive services, 3Rs are ttying as shown in Scenario 2. Even if JRs decreasethc to improve the qualityof thcir servicesin order to averageline haul, it may have a mninoreffect on seare more patronage.It is worth examiningto enployment.Theelasticiyofdecreasesinnetwork what cxtent this kind of intramodal competition is about 0.2 for both Honshu and 3-islandJRs as (whichis calledbenchnark competitionherefier) indicated in Scenario3. IfJRs inaease contracting has been intensifiedsince the privatizationofJNR. out for maintenanceofirfrastruure about 10% of 5.22 From the standpoint of the regulatory the currentlevl, they couldreduce their employees authority and shareholdersof JRs, a comparison by 1% in the caseof Honshu JRs and 1.6% in the between the peformance of JRs and the average 3-islands JRs. Although it is difficult to assess performanceof the privateraihlays can makeit easy direcdy firm this simulation how network re- to monitor thc behavior of JRs. For exanple, to trenchmnet and contractng out scheme might permit a rise in fires, the regulator can use the have positiveeffcts on productvity, the results average performanceof the railwayssector as a may indicate that their contribution to the im- 'yardstick' or a 'benchmark,' because it is outside provement of JRs' labor productivity could be the control of any single railway.Although the smallerthan we expect.This means that thcreis a decailedprocedures fr approvingfare rises is not need to examinchow such factorsas the automa- made public in Japan,it maybesafe to assumcthat tion of idcketngand ticket gate, mechanizationof the Ministy of Transportation would take into track maintenance and other technologiesmay account the averageperfomrnce of the railway influenceproduaivity. sectors,since officialstend to place much impor- BenchmarkCompetitionamongJRsand.. tance on equality of fires among the railways. ]Rs,and Prime R Enhancingcompetition based on the industy av- erage of performance is expecd to provide an The Conceptrrat Framework of Bhmark incentiveto cfficiencyand toensure that the result- Gntn al Framework of Benebmark ing benefitsare largly passed on to customers. 5.23 The originalconcept of yardstickcompe- 5.20 The privatizationof JNR differsin two tition as aregulatoryschemewasformalized first by unique rcspectsfrom a other privatizationswhich Shleifer[1985].The"cost-of-servic"regulationon have so fir been achievedor arc planned, in Euro- pricesofmonopoliescannotaddresstheproblemof pcan nations as wcll as in developingcountries, lack ofincentivesfor efficiency,since it alows that First,privatizedpassengerJRs in the main islan of 'prices track costs."There is a need tc cnsurethat Japan (Honshu) are stiUconsidered to bc regional the enrityisgivenon incentiveto engageactivelyin natural monopolies.Thanks to high population cost reduction and innovation. Shleicr indicates densityin the coastl regionsofHonshu and xocs that comparing present or past performance of siveconcntrationofeconomicandsocialactivities similar regulatedcntities could serveas a simple 82 Japanese National RailwaysPrivatization Study Il and usefulyardstidcorbenchmarkto cvaluatewhat are operatingparallel serviceswith private railways. the costsought to be and to set the right prices." This alsohas positiveimpact on effortsto improve 5.24 The comparison of cost performance the ovcaallquality of serviceof JRs. ,tmong similarentities is used in reimbursement 5.28 Althoughbenchmark competition which system that compensatcshospitals using average hasbeenobservedamongJRsandberweenJRsand costs in the USA. A similar system is used by private railwaysin commutermarkets is different Japanese insurance companies to reimburse the fromyardsticccompetition inaShleiferiansense, costs of medicaltreatmenc to hospitals.If price is this kind of intramodal competition has gready determinedby the averagecost of the comparable contributedto upgradingthe qualityof service and firms, the firm has strong incentiveto be efficient, tO holdingdown the pressurefor increasingfares in since it can "pocket"the differencebetwem costs Japan. For example,since on some local linesJRs and prices.Moreover, the regulatordoes not need are allowed to adopt specialdiscount fares that to haveexactinformationabourcostsanddemands make it difficult for private railwaysto raisetheir in order to control prices. fares, a rival could use improvement in service 5.25 Yardstickregulation was proposed in the quality such as increased frequency,speed, and LittlechildReport [ 1986]for regulationof the UK others as a mnarketingstrategy. water industry.As suggcstedin the Report, since 5.29 Interestingly,regional break-up has pro- the yardstickor benchmarkis determined not by vided an incentivefor eachJR to engagein quality- firm's own decisionsbut by the other firms' dcci- of-scrvicecompetition, because it iscasyforusers to sionsthere is no motiveto takestrateic behaviorof comparethe differencein scrvice.Also theJRwish the regulatedfirn. In addition, the regulatormay to belistedonthestocdkmarkeassoonaspossiblce not suffer fiom asymetric information about Introduction of new types of cars and through- costsprovided by the regulatedfirms, because they train servic to meet local demand are examples. norifirthe correctcOsts. It seemsthat revealing'true The most important cff ct ofthc regionalbreak-up costs'does not affectthe firm's profit-maximizing of JNR is that benchmark competition among conduct Japancscrailways has improvedthe quality of scr- OveraffAssewnentofBenehcark Gompedtion viceinurbancommuermanad aswellasininteriy mares, and the benefitsof improvementin qual- 5.26 Accordingtotheorginalconceptoflbnch- ity of serce have been passedto commutr niarkcompecition,it might be expectedthatwhen S-eies baic modelis as follows Ther aren iden-a JRs and private aihwayswhich are operatingparal- ' irm perating incain irnent.Eah nfirm es Id linesare competingwith each other for large demandcrveQ(P) mn itsmarker Whenafirm spendszon marketshareandfiaresaredeweminedbytheindus- cosc-reducnginnovaonbitanachicveunkcasckvddz). try averagecst, both of them would try to reduce wih c(O)-c.The lump sum mbsidyT istransferred to the costs ifthey could take the diflirence between fires firmif icengages in innovaton.Profit is determinedby and costs as profit. However, both JRs and the IfP-lszdi Q(-Z.T railwaysareundrnotonly stron puiiicIf de sodialwelfiare objeative is the maximizationoftwad private raways arc not only under strong publk suupluseshn theeoptimumundertheconstmintofnoloss pressureto reducefires but alsoarc not allowedto has pockt the diffen between fiuresand costs as Pc(Z) profit. It is a genral undertnding that if they are c(z)Q(l.l also to earn extra profit, this should be injected to and z-T. In sum,firsr, price is equaltO marginal cosr, eond, co- reduce fis. Under thiskind ofenvironment, they innovationmeducingoccurs up tO the poimtmatial tend to give priority on quality-of-servicecompeti- cost(-I) equalsmargna benefit;and third. the costof tion in order to gain more patronagr. innorion issubsidizedby thelump-sum ransfer. 5.27 In fact, jRs as new privatized railwayshave To diminatthe dependenceofthefirm's price on itsown costlavk theregulatorcan use costlvcls of idenical firms been attempting to set their level of scrvicedose to tw determinethe pricr. Averagecose levl c and aver that of private riways. As a result, the diflrence expenditurefor cost-rductioninnovuion z can be ob- betw=eJRsandpri- aterailwas in termsofqualky minedby dividingtotal cuss and total xpenditresof n ofservice are beosirang smaller inaccordance with fi h (n-1).The indutry averAgecost and industry d,cincn avenchpendir canbeswredasbenchmadcincompe- rheznrensayofbencbmarkcomon.TmonTtebenci- tion.What the relator should do isto commithimself nmrk competition has enhanced the quality of to the price and tasfer rulegiven by price -industy service,parcularlyoncommudnglineswhe :eJRs serg cst andTsZ7 See Vcrs andYarw [1988]. Japancse National RailwaysPrivatization Study I1 83 5.30 To assess ihegeneral effect of benchmark twcenJRs and private railwaysare not taken into competitionamong JRs and benweenJRs and pri- account explicidy,for the sakeofsimplicity of the vate railways,the question of how much scrvice analysis.Although it is desirablcto consider the quality has improved since privatizationis exam- diffcrencesin the track capacity for increasing ined herc. Thc hypothesis is that if benchmark frequencyanddhenumberofcarsinatrain,we did compecitionin certainmarkets is strong,frequency not considerthern, owing t data problems. at peak timewillincrease and the load fiLaorwill be 5.31 ThefollowingTablkV-7showsthdchanges reduced in and to win morepassengers. Two types in frequencyat peak timesin competitiveand less of raihvaylines are selcted and analyzedto obtain competitivemarkets from 1975to 1992. Generally a generalpicture of benchmarkcompectition: the speaking, there has becn sharp increases in fre- first type is the lineswhich are locatedclose to each quencyofJRsin competitivemarkets. On the otber other. The assumptionis that the shortcr the gco- hand, in less competitivemarkets no remarkable graphicaldistance between two railwaysthe stron- changeswere observed.In fact, the increasingrate get benchmarkcompetition will be and as a result of frequency of JRs from 1975 to 1992 in less greater the improvementon the quality of service. competitivemarkets is only 19%. In contrast to We havc examined Kansai-Lineof JR Ccntral, this, the frequency of JRs in vely competitive Tokaido-Line of JR Central between Nagoya markesincreasedmorethantwiceduringthesame and Osaka, Tokaido-Line of JR West, and period. The fact that the increasing ratesof fre- Hanwa-LineofJRWest. The second type islines quencyin the privaterailways are almostzero may which seem to be less competitive owing to the reflectthe situation that private railwayshave ut- lack of competitive railway lines nearby. Ex- lized capacitiesat maximum rate so that therc amples are Tokaido-Line of JR East, Ring-Line wouldbe no room to furtherincrease frequencies in in Tokyo (Yamatc-Line),Northeast Commur- those markets.In this sensc,the situationwith the ing Line (Joban Line) which connects between reprd to incrcase in frequency, for example,is Tokyo and thesuburban cities,and Ring-Linein advantagcousto JRs, since the tracks which JRs Osaka (Kanjo-Line).In these casesa major com- have inherited from JNR could be used for heavy peritor to railways is automobiles. Such freight trains. This renders it easier for JRs to intermodal competition between railways and increasefrequency and the number of trains than automobiles and also transfer of passengers be- those for privaterailways.

TalbkV-7: Changes in Servce Quaity(Frequency at Peak Hour) Degee of Ratio Competition Line 1975 1985 1992 (1975/1992) Weak Tokaido:JR East 12 17 19 1.58 Yamte JR Est 23 24 24 1.04 Joban:JREast 18 22 22 1.22 Kanjo:JR West 20 18 18 0.90 Average 1.19 Strong Tokaido:JR Central 6 7 13 2.17 (Meitsu) (25) (26) (26) (1.04) Kansai:JR Central 2 4 7 3.50 (!teitetsu) (18) (18) (18) (1.00) Tokaido:JR West 19 24 28 1.47 (Hankyu) (25) (26) (26) (1.04) (Hashin) (25) (26) (28) (1.12) Hanwa4JRWest 10 12 15 1.50 (Nank) (22) (22) (21) (0.95) Average 2.16 (1.03) (Now);Thenumbersarthenumberof trinsagpeak tiu. Thosein theparecahetsathe freque.yoftheprivate rilways whidh awrcompeting vwith JRs in the markem 84 JapaneseNational RailwaysPrivatization Study 11 5.32 As is shown in the followingTable V-8, selected from central part of Japan. The lines are the same approach was adopted to evaluate the typicalcommuter railwayswhich connectthe cen- effectof benchmarkcompetition on the load fac- tralbusinessdistrictofalargecitywithneighboring tor. We could observea similar improvement in citiesas dormitory towns.JRs have been attempt- terms ofthc load factor in the competitivemarkets, ingto keep faresrelatively low, to improveconges- although the diffcrenc bevweenvery competitive tion by upgrading cars, and to increase train marketsand lesscompetitive markets is smallcom- frequencyin order to win more passengers.Since pared with those for frequency.The averageload neitherJRs nor privaterailways don't wish to lose factor in Japanese railwaysis more than 200% in marketshare by taking the initiativeto raisefares, the lesscompetitive markets and about 150%oeven farestend to beverycomparable and stable between in the compectitivemarkets, which isextraordinazy them. Since there has been no explicitprice wars high by Westem standards.Congestion in com- between them, benchmarkcompetition in Japan muter trains isthe most serioussocial problem that has taken place mosdy in improvement of service Japamimust tacle. Since the overconcentrationof quality. population in major cities, especialy in Tokyo 5.34 Tostudytheeffectofbenchmarkcompe- Metropolitanareas, is the major causefor conges- tition on a particular JR!s market share, it is as- tion, the increasein capacityof commuter trains sumed that the number of JR's passengersmay be cannot be an answer to the problem. To solvethe determinedjoindy by the relativelevel of its Eres, congcstionproblem, there is a need to introduce loadfactor,and trainfrequencyincomparisonwith damsticland-policy refonns to reducethe concentra- those for the privaterailways. JRs since privaiza- tion of populationin Tokyoand to chan land use. tion appear to bave used the lcvelof servicequality Tbe Effca o Benchrby the privaterailways as a benchmarkto establish Ui Beca Cheir markeing suargics.Since passenger chooses Marherailwm wrich offers rlatively lower faresand 5.33 Toassesshowbenchmarkcompetitionh higher qualityof services,there woult be a direct influencedthe marker shareofJRs in terms of the cause-effectrelationship between the number of number of passenger,threc commuterlines wbere passengerandtherelativelvclsoffares,loadfactor, JR and privaterailays are operatingin pallel are and train frequency. It is obvious that train fre-

TableV-. Changesin Sence Qualily aud Fator at PeakHour) (Paed) Degree of Ratio Competition Case of Lines 1975 1985 1992 (1975/1992) Weak Tokaido: JR East 263 256 226 0.84 Yamate:JREast 246 263 272 1.11 Joban: JR East 206 259 248 1.20 Kanio:JRWest 232 244 156 0.66 Average 0.95 Strong Tokaido: JR Central 165 168 130 0.79 (Meimn) (213) (184) (168) (0.79) Kansai:JR Central 0.83 143 143 1.63 (KWimts) (200) (165) (165) (0.83) Tokaido: JR West 199 184 122 0.64 (Hankw) (209) (174) (159) (0.76) (Qansbin) (162) (162) (153) (0.74) Hanwa.JRWest 212 183 169 0.80 (2V=kd) (182) (179) (172) (0.9:' Average 0.97(0.81) (Note):Thenumnbess arethenumbrofT nsatpeakumcTose in theparenthesesaetic fzequecyofthe preaiways,whidc arccompcong with JRs in dhemarks Japanese National RailwaysPrivatization Study II 85 quencyisaveryimportantdemcntfortheselccrion comDaredwith that for the private railways.The of the railwayin the commuter mnarketsat peak rario one indicatesthe 50-50 skarebetween the JR time, becauseJapanese commuters as wel as those and the privaterailways. Thc horizontalaxis shows in other countries are likelyto place priority on the JR's rclativcmagnitude of the three strategic reachingtheir workplaceson timc. Althoughthere dements such as fares, frequency, and the load areotherimporamnt factors to influencingthechoice factor to thoseof the rival privaterailways. Again of commuters,only thrce elementsare considered the ratio one meansthat thescthree dementsarc in herc becauseother data are not available. balance among themLIncreases in JR's competi- 5.35 Theconceptualframeworkofanalysisused tivenessrequire movementsof the share-strategic here isas follows:IfaJRwishes to expandits market elementsprofile in the northeasterlydirection. The share,it would ty to lowerthe relativefare, increase following areexamples which show how bench- the rlative frequency,and reducethe loadfactor in mark competition is working. Since we selcted comparisonwith thoseof rivalprivate railways. In lineswith the samedismnce, fares in the following the competitivelinc, the railwaywith the better Tablesarconcswhicheachrailwaycompanydcargcs. qualityofserviceas well.s lowerfarecs will be likely 5.36 XlineCase:AsshowninTableV-9,theJR to carry a larger-than-proporrionateshare of the sinceprivatzation has becn increasingfrequencies traffic.Becauseofferingbetterservices isbelieved to at peak time Although the private railwaystill cnablerailways to capture the relativelarg propor- keepsits dominantmarket sharein this line, thcJR tion ofavailabletaffic, each railwaywouldattempt has been successfiulfor obtaining more patronage to gain an advantageby increasingfrequency and than beforcand expandingsignificandy its rmiarket lowering load factor and being a followernot a sharefrom5.7%in 1987to 11.31%in1991.Since lcader in price wars. Graphicallythis relationship the difrence in faresbetween the JR and privatc forms as S-shared airvc as in Figure V-2. The railwayshas ben widrin thbismay have some vertical axis shows the relative share of the JR inftuencontheexpansionofpassgausingtJR.

FigureV-2 TheConceptual Chart of Relationshipbeten the MarketShare and Strtegic Elemens

100% i_

Relative Share

Retive Fares(or Frequency ar LoadFactor) Not: The verticalaxis = The relativeshare of JR to tat for the rival privaterailways Thehorizonrmlaxis=Thcsrec elementsuchasthe tioofR's Ers, fiequency. and theload facor tohoseofdle privareraihways Increase in JR:scompetitiveness require movements of the sc-straegic eiementsprofik in the northeastey diection.

TableV-9 BenchmarkCompetion in X Line 1987 1988 1989 1990 1991 Number of passengerat JR 1,214 1,653 1,230 2,391 2,680 pek- hour Private 20,050 19,510 19,890 20,430 20,450 Fare JR 440 440 450 450 450 (yen) Private 480 480 490 490 540 Load fiiaor JR 139 140 118 131 143 at peak(%) Privamte 162 158 161 165 165 Frcquencyat peak hour JR 4 5 7 7 7 (no. of taains) Private 18 18 18 18 18 86 JapaneseNational RailwaysPrivatization Study II 5.37 TheFigurcsV-3,4,5showthcrelationship increasein its relative share of passengerson this betweentheJR'smarketshareandtherelativefares, line from 1987 to 1991. load factor, and train frequencyof JR to those for 5.38 Y Line Case: In the caseof Y line, the JR the rival railway.Obviously the movementsin the had rdatively a large share of the market before northcasterlydircaion showimprovement ofJR's privatization.The JR aftcr privatizationmade ef- competitiveness. Althoughno one singlefactor is fortsto increasethe frequencyoftrainsat peakltime decisivein improvementsinJR'scompetitive edge, by 60% from 1987 to 1991. In contrast to this, the it seemsthat relativelylower fares and higher train private railway did not have room to increase frequencr of JR have bEen contributing to an frequency,because of the limitcedcapacity ofinfra-

FigureV-3: The relatonship between therelatve levelof JR's faresand JR's marketshare

FAREVSSHARE

0.14 0.12 .--. 0. 1 | 0.08. - 0.06- - 0.04 0.02

1.05 1.1 1.15 1.2

Note: Thevercal axis= the marketshare ofJR in cheX linefiom 1987 to 1991. the horizontal axis = the relative advantage ofJR's fes against the rival in the X linc fiom 1987 to 1991.

FigureV4 Therelationship betmeen the relaivelvel of JR'sload fator andJRfs maket share

LOADFACORVSSR4IRE

0.14_ 0.121 0.1 + 0.06+\ 0.08- - 0.04 - 0.02. 0 0.s 1 1.5

Note: The verical axis = dte market share ofJR yaginsr the riv2l in dte X line fiom 1987 to 1991. The horizontal axis shows the relative advantge ofJR's load factor aginst the rival in dte X line from 1987 to 1991.

FigureV.5: The relationshipbetween the relative klve of JR's trainfrequency and JR's narket share

FJNCYVSSHAE

0.14 0.12 * 0.1 0.08- 0.06 - 0.04 0.02

0 0.1 0.2 0.3 0.4

Note: Theverical axis =the market share ofJR in theXline fmm 1987 to 1991. Thehorizonau axis = the relativeadvanrage of JRs train frequencqagainsr the rival in dieX line fiom 1987 to 1991. Japanese National RailwaysPrivatization Study II 87 structure.As a result, JR has succeededin expand- have provided favorableconditions for the JR to ing its market share from 19% to 28%. Because regainits narketshare.Note thatdueto changesin fireswere in balancebewecn theJRand the privatc the location of taffic reseach from 1989, train railways,the competitiontook dheform of increas- frequenciesand load factorsin 1987 and 1988 are ing frequencyand reducingthe load factor.Table adjustedto maintain consistencyof data. V-10 indicaes rhestraregicpoliciesofbothJR and 5.41 In Z line the rdative low fiLresof JR and the privaterailway in competinglines increascin train frequencymay be rcsponsiblefor 5.39 The followingfigures indicate how ixn- the increasesin JR's market share from 1987 to provementsinJR'sservicclevelsmayberesponsible 1991.The foiowing FiguresV-9,1 0, and 11 show forthe increaseinpassengersfortheJR inYlinefrom the relationshipbetween JR's market share and 1987o 1991.BoththeJRandtheprivaterailwaydid three variables.In this line there is a clear-cut not raise firs for six years,owing to bendmuark relationshipbetween the marketshare and reativc competition.However, the JR's marketshare has fires as wellas relativefrequency, but not betwen increaseddristcaly. The FigureV-6, 7 and 8 may the marketshare and relativeload factor. indicatethar when faresare not diflint between 5.42 Therc is no single decisivefator for in- rivals,the relativeleel of sevice qualiy becomes creasesin the reladveshare of JR's passengerin imporant t determinerelative maret shares.In this dtese three lines, becausevarious elements joindy line therelative increas- in train frequencyseems be havecontributed to improvementsin JR's relative a kry for the increase, -R's marketshare. competitiveness.However, the load factor is obvi- 5.40 ZLine Case:As LabklV-1 I shows,JR.had ously not an important factor for measuringthe a relativehigh share in thismarker and expandedits relativeincrease in JR's market share, pardy be- markershareafterprivatizationfiom 1987to 1988. causethe load fictor might be used as a meansfor However,the privat railwaycounterattacked JPs increasingtanisportation capacity for the short expansionpolicy by increasingits frequencyfrom terms in order to cope with the shotage of the 1989 to 1991.As a result,JRwas forcedto reduce capacity,and pardybecause the load facor itselfis its marker share compared to the private railway estimatedbydividing the number ofthe passenger from 1988 to 1989. JR has graduallyrgained its bytrain fequencyatpeaktimc. Theecasestudies competitivenessfiom l990 onwardswhenthe rival's indicatetbatbenchm arkcompetitianhas enhanced concernraiscd its fuares.JRs lowerload faacors may the levelofservicc quality in thcsccompetitive lines

Tabk V-10Benchmark Competton in YLUne 1987 1988 1989 1990 1991 Peakno. JR 8,876 9,899 12,871 13,013 14,706 passcngers Privatc 36,829 37.127 36.815 36.970 37,170 Farc(n) JR 430 430 440 440 440 Private 430 430 440 440 440 Peakload tcor JR 174 167 134 124 130 (%) Private 181 178 171 169 168 Peak fequency JR 8 10 10 11 13 (no. of trains) Privatc 26 26 26 26 26

FigureV-6: The relationship between the relative level of JR's aes andJR's market share FAFEVSSHAE

0.4-U 0.35 a 0.3. 0.25. 0.2 . 0.15-. 0.1 0.05

0 0.2 0.4 0.6 0.8 1

NotcThveti axis= themarketshofjR. Thc horzon axis= th elaiveadvantacofJKs fawsagainstherin heYline from 1987 to 1991. 88 Japanese National RailwaysPrivatization Study II and thebenefits of improvemcntshave been passed gersmay bc determinedjoindy by the relativelevel on to commuters. of fares,relative levelof load facor, and relative Appendix A Simple Integrated Analysis of eveloffrequency.Notethat,economricaiyspeak- BenchmaukCompetition ing,because the load factor is detcrminedby divid- ing the number of passengersby train frequency, 5.43 It isdifficlt to analyzedirecdy the overall there mightbc a strongtendency to correlatetraffic effcct of benchmark competitionon the modal with thesetwo fictors. Althoughthere is no wayto choicebecuse data are limited to only five years avoid these statisical problems, since there has afterprivatzation. We couldnot applyany econo- been no attempt so far to investigatethe effectof metricmethods to study the e$ffctsof benchark benchmarkcompetition on marketshare, we have coMpetitiononmarketshares. Thefolowinganaly- tried co adopt the followingsimple analysis. It is sis isan attemptto assesshowJR'sefiTrts mayaffcc assumedthat there is a linear relationshipbetwecn its maket sharethrough benchmak compeition. JRls relative level of fares, load factor, and fre- Thcbasicassumptionis that the number ofpassen- quency and JR!srelative market shareas follows:

F1i V.7:The rdeaonship betmn th rdeaveeve of JR's oadactor and JR's mabtshae

LOADFWTRVS94VE

0.4 . 0.35 _ 0.3- 0.25 - 0.2 0.15 0.1 0.05

0 0.5 1 1.5 Note: 'e verticalaxis - die m- shareofR The horizontl axis- the relativeadvantge of]Rs loadfiaor againstthe rival in the Yline fiom 1987to 19,91. Fiw V-8The reltfomnip bt n thereat ev of JR'sftin fruncy and JR'snmaket sham

0.354. 0.3 - 0.25U 0.2 0.15 0.1 0.05 0 I I I 0 0.1 0.2 0.3 0.4 0.5 NotG lhevercal axis- he maeshareofJR Thehoizontal is - the datveadvantagerofJRsuain fiequencyagainst the rivl intheYlinefiom 1987 w 1991. Tabl V-1lBenchmrk Competiton hi Z Une 1987 1988 1989 1990 1991 Peak no. of JR 9,515 10,515 10,117 10,933 12,404 pasengs Pivate 29,574 29,768 30,290 31,075 32,259 Fare (V) JR 380 380 390 390 390 Private 390 390 400 460 460 Load fiaor JR 176 169 126 122 130 at peak(%) Private 173 171 166 162 159 Peak fiequency JR 7 9 9 10 I 1 (no. of trains)Private 27 27 28 30 32 JapaneseNational RailwaysPrivatization Study II 89 PjdPpv=CI(Fpv/Fjr)m+(Lpv/Ljr)b+(Qpv/Qjr)cI The larger the value of C is, dhesteeper the slope where becomes.C maysuggest how changesin JR's fares, PjdPpv = the ratio of JR's passengerto the load factor, and frequency could affict on the privaterihvay's passenger. modal choiceof commutenswho might selectthe Fpv/Fjr= the ratio of faresof the private rail- railwaywhic offirs bcter qualityof service,other waysto JR's fares, thingsbeing equaL In that senseC could bea proxy Lpv/Ljr= the ratio of load factor of the private to show the sensitivityof conunuter choice. C is railwaysto JR's load factor, calledchoice index hereafier, and showsindirecdy Qpv/Qjr = the ratio of train frequencyof the the intensityofbenchmark competition over tume. privaterailways to JR!s trainfrequenqc. In reality, JR has been trying to keep fares un- 5.44 TheequaEionindcatethatifthesituation changed, reduce tde load factor,and increasefri- is favorableto JR in tems of the relativeratio of quency to win more passengers.Although the fares,load factor, and frequency,JR's relativemar- analysiscannot rakeinto considration how many ket share will increaseover rime. C is a kind of passengershave shifredfromaprivae railwk roJR coefficientto determinethe slopeof the equationL and viceversa or from passager casnto JR. it may

RigureV-9: The readonsNp befte te relatie levelofJR's faes andJR's nakt she FAREV3SSARE

0.4. -

0.3 - 0.2- 0.t. . 04 1 1.05 1*.1 1.15 1.2 Noe Theverial axis- dte madkrsham ofJRin dheZ finefrom 1987 to 1991.Th horizontalais - thereative advanag of JRfas againstdie rivalin de Z linefiom 1987 to 1991. RFureV-10 The reons p betwenthe reai lvl of JR'sload torand JR'S nt shar LOADFACMVRVUSSHARE

0.3 1 0.2-

0.5 t 1.5

Note Theverticl axis= the nma sare ofJR in dte Z line fiom 1987 wo1991. The horiwonmlaxis . de rdati advantageof JR!sload fcaor againse die rivalin heZ finefiom 1987to 1991. FigureV-11 The relationship between the r i lvel of JR'stin rquencyand JR'snwket FREQUENCYVSSHARE 0.4- 0.3 .

0.2. 0.1. .

0t 0.3 0.35 0.4 Noe Thevetical axis = themarket sha ofJR in de Z lne from1987 to 1991.The horiontl adis the relativeadvanuge of JR9strain frquency against die rival in dieZ linefrom 1987 w 1991. 90 Japanese National RailwaysPrivatization Study II show the intecraaionbetween JR and a privace tive becauseifJR's relativefrequency is largerthan railgay. that for the privaterailwaysjRs could increase their 5.45 Althoughthe coefficients.a, b, and c are marketshare, which is shownas the inverseof the unknown,itisobviousthataandbare non negative relative frequencyof JRs to the private railways. values (a, b > 0) and c is a negativevalue (c < 0). Although there is a need to estimate these coeffi- Since a and b are the coefficients to show the cients (elasticities),because of data availabilitywe rdative levelof faresand load factor of JR to the did not estimarethem. For the sake of simplicity, private railways,their positivevalues mean that if we define here a and b as being plus one (a,b = 1) JRs ofrer relativelylower fares and lower load and c is minus one (c = -1). If JR and dte private factorsthan their counterparts,JRs could increase railvay share the market equally, and if JR is their marketshare The coefficientc must be nega- equivalentto the privaterailways in terms of fares,

Tabe V-12Changes In ChoiceIndex in the case of X-Line 1987 1988 1989 1990 1991 C Values 0.024 0.034 0.022 0.043 0.046

Figure V-12 ChangesIn JfRs ChoiceIndex in the caseof X-Line

0.3

0.2

0.1

N 0 co 0 _ 0 CD 0 a> a

Figure V-13Changes in JA's relaffvecompetveness

C-Value

0.3t

02-

0.1

0 _ 2 2.5 3 Three StrategicElements

Note Thevertcal axis = C-valuesTWhe horizontl axis = RelativeFare + RelativeLoad Facor + RelativeFrequency Movemcnrs to the northeast dircaion mean thatJR7s r5ativc advantage is i.ceasing over ite. JapaneseNational R. ilwaysPrivatization Study II 91 load faaor, and frequenr-, C will be 1/3 or 0.33, ence toJR due to bendumarkcompeticionin X-Line becausethenumerator i).Ppvisequalco1 andthe increasesabouc twicefiom 1987to 1991.FiguweV- denominatoris the suni of Fpv/Fjr=1, Lpv/Ljr=1, 13 showsthe changesin the relatioi'shipbetween and Qpv/Qjr=l so that the sum of thcsc ratio JR'scompetitircnessand threevariablesover time. becomcsthrce. In other words,when the compcti- 5.47 In Y line, since the JR increasedits frc- tivenessofJR and the privaterailway is in cquiib- quency, it becamcmorc attractivcthan before.As rium,CwouldbcO.33. The rclativcincrease inJRs' shown in Tablc V-13 and FigurcV-14, the calcu- competitivencssowing to bendcnark competition lated C values in Y line are not only higher than can be measuredby the deviationof C from an those in X-Line but alsoincreased by about 40% equilibriumvalue of 0.33 over tim.c from 1987 to 1991. The JR has succeededin 5.46 TablesV-12 to 14 and FiguresV-1 2 to 17 cxpanding ie markersharebyimprovingthequal- show the changesin C valuesof three lines,which ity of servicc. FigurcV-15 shows changesin the may indicate the degree of the modal choice of relationshipbercwenJRscompecitivenessandthree benchmaucompetiton from 1987to 1991. Prefer- variablesover time.

TableV-13 Changes in ChoiceIndex in Y-Line

1987 1988 1989 1990 1991 C Values 0 103 0.109 0.132 0.126 0.142

RigureV-14 Changes in ChoiceIndex in Y-Line

0.3-

0.2 -

0.1 - -

co co coo im - 0 m CD 0 CD

Fgure V-15Changes inJR's Re/adveCompees CValue

02-

0.1 .

2 2.5 3 Three StrategicElememts 92 JapaneseNational RailwaysPrivatization Study II

5.48 The followingTable V-14 and FigureV- competitivemarkct, if it was freed from the heavy 16 show the changes in Choice Index in Z-Line burden of maintaininginfrastructure. Foz the first from 1987 to 1991. Since the private railway twoorthreeyearsafterseparationfromJRspassenger, continued to inciease its frequencyfrom 1988 to dtanksto theemonomic boom in the late 1980%anid 1989,JR's competiriveedgewas reduced, but after the severeshortage of workersin the trding indus- owingto incrase in the relativefiequency ofJR, the uyJRFreightmadefairlygoodprogressinexpanding situationforJRgradualy has been improving Figture its market share. However,af;er the burst of the V-17 showsthe changesin the relationshipbctrw:cn bubbleeoonomyin spring 1991demandsfor railway JRlscompetitiveness and threevariables over time. freighthavebeenfallenrapidly,sineeshippinggeneral EconomicAnalysis ofJR Freight goods by railwayis consideredthe last choicefor customerswho place prioniy on deliverytime. As a Cbnge in Costand :Serice Ogputpxt efiRFirit rcsult,JRfreightfinds it difficut to maintaineven the 5.49 The main purposes of the creation of a currcntrmarket share, which isless than 5% of the nation-widcJapan Freight RailwayCompany JR total freight market on a ton-kn basis. Freight)werc to reduce rhesnowballing deficits of 5.50 Onc of the problems JR fitcesis hcavy the frcightsection ofJNR and providean opportu- dependenceofJR Freighton forwardersto collect nity for rcstructring freight operations and re- freight. Traditionally JR Freight has not been dressingineffiiency.Itvasexpeced thatJRFrcight equipped with the necessarymarketing functions couldstandalonefinancialiyandsurvive in thevery and does not havctrained people to engagein sales

TableV-14 Changesh ChoiceIndexin Z-Line 1987 1988 1989 1990 1991 CValues 0.142 0.149 0.125 0.124 0.140

FigureV-16 Changes in ChoiceIndex in Z-Line

0.3-

0.2- 0.1A~*Ew |

0- co CD co o CD

FigureV-17 Changes in JR'scompetitveness

C V_ue 0.3

02

0.1

0 0 t 2 3 Three Strategic Elemets

Note: The vertical axis= C-values The horizontalacis = Reative Fare RelativeLoad Facor + RelativeFeqtiency Movementsto die northeastdirection mnean that JRs relativeadvantage i: increasingover time. JapaneseNational RailwaysPrivatization Study Il 93 aaivities. JR Frcight is in a very weak positionto 5.51 JR Freight in cflcct borrows tracks by accessdirectly its dients in the generalgoods mar- paying user feesto JR's passengercompanies. Un- ket when comparedwith forwarders.The trucking der thissituation,the imporrantquesrionis whether companiesare doser to final consumersthan JR or not the user fees JR Frcight pays cover the Frcight and have better knowledgeof what the relevantcost oftra cuse.According to the contrat customerswant. Trucking companieswere able to berweenJRs passengerand JR Freight,user feesarc make use of the transportation capacity of JR supposed to be basedon avoidablecost. To asscss Freightas a bufferto adjustsupply and demandfor whetherJR Freightfinancially scands alone or not, their own interest.When they cannot handle de- it is importnt to study whetheruser feesare dose mand, for example,because of a shortageofdrivers to the marginal or averagecost of railway opera- or trucks, thry may use JR Freight to transport tions. If the user fee is smallerthan the marginal goods. Since there is asymmetricinformation be- cost, there will be a possibilityof cross-subsidiza- tweenJR Freightas principaland truclcingcompa- tion betweenpassenger JRs and JR Frcight. If so, nies as agents,JR Freightcan monitor imperfctly the user fee JR Frcight currendypays may not be the opportunisticbehavior of trucing companies justifiedfrom the standpointof efficiency. who mayact cooperativelyor non cooperativelyto 5.52 FigureV-18 showsdcanges in ton-km of mazximizetheirownprofits.TheMinistyofTrans- railwayfreighEfrom 1970 to 1991. Railwayfreighr port attempts to promotc use of railwaysin the showeda sharp dcdine inhe 1970sand 1980s,but freightmarket from the viewpointof coordination sinceprivamaonin 1987denandfor railwayfreight among transportation modes. But JR Frcight is has shwm an ovall incacasingtrend. As shownin losingits competitiveedge mainly because oftimc- FigureV-19, howcver,demand began to dcase cost inefficicncy. againfrom 1992, owing mainly to the sluggish F,gureV. 18: Trendof FreightTransport (Ton-kmn X=yearY=ton-km (billion)

60

0 - I I I 40s- 1

2- 30 75 80 e, 90 95 FigureV-19: Trend of OperatngCost hess Depreion) X=year Y=operaingCost (less depreciaton) (billion yen) 4200.

4000. J ______3800. - 36°o'~~~~~~~~ I U 3600-

3400 - ______3200

3000. - ______2800-

2200 70 75 so 85 g0 95 94 Japanese National Railways Privatization Study II economy.Inspitcofdecreasingdemands,the operar- structurc. Thirdly, avoidablc costs are calculated ing cost ofJR Freight surprisinglyincreased sharply. undervariousscenarios. Finally,theesuimated avoid- The opeating costs readced peak at 1985, and siwce able costs are compared with the average cost and then there havc becn paralle changa betwee de- marginal cost in ordcr to decermine whether thc mands and cosL This mc;ns that during the relevant user fee for the track usc is appropriate or not. This period theJNRfreightscionmayhavebern aniajor maygive an indication as to what extent JR Freight cause for the JNR's huge deficit. To restructure the is aoss-subsidized by JRs passenger. dcficit-ridden freight section, JR Freight necededto 5.55 The avoidable costs aredefined as foliows: bc cut offfiom passenger transportation activities. Suppose that C(Qp,O) is Sand-Alonecost rooper- Ezsmation of Cost Fion te only passenger railways(the cost of supplying palssengerscrvicein isolation),andC(O,Q) isStand- 5.53 JRs as a whole joindy supply two different Alone cost to operate only freight railways(dte cost services - passengcr and freight - using the same of supplying passenger service in isolation). Ct is infrastructure. Since JR passenger companies own the sum of two Stand-Alonc cost, that is Ct = the track,JRFreight is supposed to pay user fees for C(Qp,O) + C(O,QfO.Assumetat Cj = C(Qp,QO is the use of track based on the avoidable costs, as the joint cOst when a railway company supplies discussed earlier. The problem is how to allocate both passenger and freight services at the samc common costs which indude not only the cost of time. Since the avoidabl costs are defined as csts providing track but also the costs for signaling savedif either passengerscrvice or freight service is terminals, electricity generation stations, and so withdrawn, they are obtained by dividing incre- forth, since conmnon costs forxn a very substantial mental costs by output as follows: part oftl costCfor railwayoperations. Moreover, AVCp = IC(Qp,Q) - C(O, Qf)l / Qp, for there isalargeportionofaommoncostwhich isnot fixed Qf cearly traceable. JR passenger companies use an and accounting method that allocates cormon costs to AVCf= IC(Qp,Qf) - C(Qp,O)l I Qf, for oneacivityoranother. Froman economicpointof fixed Qp view, however, accounting costs are of limited whereQfstandsforoutputoffreightserviceand guidance in setting user fees for the use of track Qp shows output of passenger service. There are Generly speaking, the accounting method does twc kinds ofthe avoidable costs. The problem is in not conform to economiccosting methods. The defining the base from which to begin to the scries problem is that there are significant differences of allocations, passenger service or fright service. associatedwith the costs of track used cxdusivdly Which definiuion is used depends on whethcr the forpassaergrservicesand that usedonlyfor freight. infrastructure of railway is mainly designed for This means that the calculation ofavoidable cost is passenger service with freight service imposing changed by its dcfinitions. The concept of avoid- additional costs or not. If priority is given to able cs becomes rather dubious as a mean to passenger service, AVCfshould be used. lfpriority allocate common costs. It is important to study is given to frcight service, AVCp should be used. whvethrJR Freight is covring its specificcosts plus However, it is a general understanding in Japan contribution to overheads from the standpoint of that passengerjRs provide important intercity ser- economic analysis, because if user fees forthe track vices and also play a key role of providing com- usearelessthan theavoidablecosts, thenJRFreight muter services in major urban areas. It may be is making a loss for JR passenger companies. reasonable to suppose thatJR passenger companies 5.54 To determine whether JR Freight may have priority over JR Freight. In this case, the cover the rlevant track costs fiom the viewpoint of avoidable costs are calculated on the basis of how cfficincy, the following procedure is used here: much JR passenger companies could savc cost of first, operating costs of passenger and freight ser- operations if frcight servicrs were abolished.The vices are estimated by Eimz-seriesand cross-sccion avoidable costs can be daetrmined by the joint cost regressionanalyses using data of output of pasen- C(Qp,Qf) minus cost of passenger service C(Qp, ger and freiightand network conditions. Secondly, 0), that is AVCf. the aveage cost and marginal cost, economies of 5.56 To judge whether user fees are relevant or network density, economies of scale, and econo- not, there is a need to know whether economies of mies of scope are calculated to dearij the cost scopeandscaleexistinJRsoperationsasawholc. If JapaneseNational RailwaysPrivatization Study II 95 the joint cost (Cj) issmaller than the sum of Stand- Qf = output in freight transport ( ton-kIn) Alonccosts (Cc), this may indicate the existance Ni = nctworkvariables (i.e. route-klm,trip economies of scopc in railway operations. In length ctc.), and D = dummy variables otherwords, if the total costsofsupplying jointly (ic. privatizationdummy). passengerand freight servicesby monopoly arc 5.59 Since samplesare small,simple log-linear smallcrthan those for supplyingthem separately cost function is used here. The advantagcof log- by two independent railway companies, this lincarcost function is that the coefficientsof each may indicate that the joint production of pas- variablerepresent the elasticitiesof cost. To avoid senger and freight servicescan save costs com- biasofestimation, difrent sample setsand differ- pared with separable provisions of these thcse ct kindsof modelsare introduced.Basically two services. If the rcsult is the opposite and so kindsof cost specificationare usedas folows diseconomies of scope exist, the break-up of (for time-seriessample) passenger service from freight service or vice ln(C) = a + bt* ln(Qp) + b2* ln(QL)+ c* ln(N) versa may be desirable. Our measurement of + d* PRV economiesof scope (ESP) is the ratio of the sum (for cross-sectionalsample) t of Stand-Alone costs to total costs as follows: ln(C) = a + b1 ln(Qp) + bz* In(Qt9 + l; c* Economies of Scope (ESC) = [C(Qp,O) + ln(N;) + d*JR C(O,Qo]/C(Qp,Qf) wherePRVand JR aredummyvariables.In the If ESC is greaer than onc, this means that first estimationequation the dummy PRV is as- economiesofscope exist and if it is lessdan one, it signeda onc for after privatizationand a zerofor showsdiseconomies of scope. before privatization.In the second equation the 5.57 Disoconomiesof scopein railway opera- dummy JR is a one for JR and a zero for private tions mightaise if thejoint use of trac for passen- wa gerand frcightservices have affeacte dverselyeach 5.60 TableV-i 5 showsthe resultsof estimation other, for example,owing to oDnflictof scheduling in cass oftime-seriesordinay last squaremethod and the need for managerialadjustment between (Model-i), first-order autoregressive method thcL Thecremight be lag transaction costs for (Moddl-2), cross-sectional method (private rail- adjusng the two different operations.The joint waysonly, Model-3),and cross-sectonalmethod 2 cost q could becomelarger than the sum of two (JRsplus privaterailways, Modcl-4). The R 's are StandAlonecosts ifeconomies ofscopearcsmaller vey high, although the t-satistics are relatively than diseconomiesof scope.Whether or not the smalLGenelly speaking,route-km has highclas- joint cost becomeslarger than the sum of Stand dcity in most models,which show that Icngth-of- Alonc costs may depend on whether or not routevariablemayhaveastrongimpactonchanges diseconomiesofscope(cost-augmentingaacivies) in cost. In cross-sectionalanalysis, clasticity of are larger than economaiesof scopc (cost-saving passenger-kInisalsolargebothinthecaseofprivate actvities). In this case,the calculationofavoidable railwaysonly and in the case of JRs plus private cost is far more complex than we expect. It is ailways,which may indicatethat output ofservice extremclydiffiaict o separateeconomics of scope is an important factor to set the cost. However, from diseconomiesof scopein railwayoperations. diffrences in the size of output betweenJRs and Insteadwe willuse regrcssionanalysis to giveover- the privaterailways arextrem ly large,which may all assessmenton the relvancy of user fee. give biasto the esimation results. Esirnion of CostFAnzeen ofiBs 5.61 Table V-16 shows the estmatedaverage cost and marginal cost of JRs using the above- 5.58 It iSassumed chat costs are explainedby mentioned models.Average cost of passengerJRs twomknds of serviceOUtpUts, networkfaors, and in time-seriesanalysis is about I1 yen per passenger- somedummy vaiables as follows: km andtat for cross-seccionanalysis is in between7 C = Cp + Cf= f (Qp, Qf, Ni, D) and 8 yen per passenger-kn.In caseof jR Freight, where averag cost mangesfrom 43 to 88 yenper ton-kmnin Cp = operang cost of paswnger the time-seriesanalysis,but in crosssecionanalysis it Cf= operatng cost of freight has hoveredaround 35 yen per con-km. Qp = output in passengertmnsport (i.e. pas- 5.62 Economiesofnetworkdensity,economies senger-km) of scale, and economicsof scope using the same 96 Japanese National RailwaysPrivatization Study II model specificaEionare shown in Table V-17. production issmaller than individualcosEs of pro- Economicsof network densicyare defined as the duccion.Thercareverylargeeconomiesofnetwork proportional increasein costs due to the propor- densiq in timeseries analysi :, alkhoughin the cases tional increasein output, fixinginput prices,net- of cross section analysis cconomies of network work faicaors,and service output characteristics. density decline almost one-half compared with Economicsof scale are calculatcdas the propor- time-scriescases. As far as cconomiesof scaleare tionalincrease in cost resultingfrom the propor- concerned,they rangefrom 1.00 to 1.58,which are tional increasein output and route length, fixing a litde higherthan North Americanrailways study input prices and service output characteristics. by Cave, Christensen, and Swanson [1981], but Economiesof scope mean that the cost of joint dosetotheresultofAmericanurbannailwaysstudyby

TableV-15: Estimaon Results: Coefficients and Standard Errors Model Model-I Model-2 Model-3 Model-4 type of time-series time-series cross-sectional cross-sectional sample (1981-91) (1981-91) (private only) UR+private) Mehod OLS ARI OIS OIS ln(Qp) 0.172 0.151 0.422 0.348 (passge-kin) (0.348) (0.391) (0.062) (0.127) ln(Q ) 0.044 0.127 0.084 0.076 (ton-kIn) (0.125) (0.161) (0.076) (0-157) ln(Lr) 0.505 0.353 0.493 0.533 (route-kn) (0.601) (0.696) (0.120) (0.h23) ln(lp) -0.545 -0.462

trip icngth) - (0.084) (0.176) In0h) -0.108 -0.102 rdei trip lngh) - - (0.090) (0.186) PRV -0.209 -0.224 - - (privatizationdummy) (0.083) (0.095) - - JR - 0.330 (JR dummy) - (0.847) constant 28.982 28.995 24.495 25.769 (0.041) (0.048) (0.232) (0.866) RS*2 0.9091 0.9203 0.9600 0.8522 Not= Thenumbers in the prnthesis amede standar errors.

TableV-16: Estmated Average Cost and Marginal Cost model-1 model-2 nodel-3 (cross- model-4 (cross- (time-series: OLS) (time-series: AR1) sectional:private only) sectionaLJR+private) ypeof COSt AC MC AC MC AC MC AC MC passengerjRs 11.5 2.0 11.1 1.7 6.9 2.9 7.9 2.8 ficightJR 43.7 1.9 87.8 11.1 34.2 2.9 35.5 2.7 Note 1)AC is avenge incrementl cost (CI-Cp)/QQ and MC is margind cost. which is obtained by AC times dasticity. 2) These numbess are estimated results using 1991 statistics ofJR.

TableV-17: Economies of NetworkDensity, Scale and Scope model-I model-2 model-3 (cross- model-4 (cross- (time-series: OLS) (time-series: ARI) sectional:private only) sectional:JR+prIvate) Economicsof netwodCdensity 4.63 3.60 1.98 2.36 Economiesof scale 1.39 1.58 1.00 1.04 Economiesof scope 0.76 0.43 0.26 0.30 Japanese National Railways Privatization Study II 97

Viton 11980]. This cost study suggCsts dut there elasticity. Although samples are small, it mav be would be diseconomies of scope in JRs, since every safe to condude that JR Freight docs nor cover values show far less than 1. avoidable costs in any case. JR Freight pays only 5.63 TheestimaEionofavoidablecostsisshown 0.65 yen as the user fec which is far lower than in Table V-18 under various scenarios. Scenario I the marginal cost cstimated here. From this indicates thaE all joint costs, in other words, all econometric analysis there is little doubt that JR inacemental costs resulting from joint production Freight is subsidizedbyJR passengercompanies. of passengerand frcight service, are allocated to JR The relationship betwcen JR passenger compa- Frcight. Scenario 2 is that all joint costs are attrib- nies and JR Freight must be changed from the utable toJRpassengercompanies. In Scenario 3, all viewpoint of efficient resources allocation. How- joint costs are allocated by share ofoutputs assum- evCr, in the definition of avoidablc costs, rel- ing the average weight of a person to be 70kg. evant costs are tied to the particular railway Scenario 4 shows the estimation of avoidable cost policy. There arc social costs associated with using JR Freight's marginal costs. The avoidable road transportation which may jusrifythe cross- cost per ton-km in this case is estimated as being subsidizationofJRFreightbyJRpasscngercompa- between 14 yen and 43.7 yen per ton-in. nies if social costs of railway freight is less than that 5.64 Table V-19 shows that the comparison for trucking. As the second-best theory suggsts, if of user fee and average and marginal cost based prices in road scaors deviate from costs, there are on time-series model-1. Since clasticity of ton- sound economic reasons for railways to deviate kmu(InQf) is 0.044 and average cost is calculated from the marginal cost pricing. There is a strong by (Ct-Cp)/Qf, we can obtain the marginal cost need to study the costs of railvay freight ranspor- of 1.9 yen per ton-km by multiplying AC by the tation from the viewpoint of externalities.

TableV-18: Estimation of AvoidableCosts avoidable avoidable cost for fieight's share cost cost per ton-km description freight only of joint cost of frei*h (V/ton-kin) Scenario 1 all joint cost is 381,314 784,991 1,166,305 43.7 allocated to fxeight Scenario 2 all joint cost is 381,314 0 381,314 14.3 allocated to passenger Scenario 3 all joint cost isallocated 381,314 476,411 857,725 32.1 allocated by share of outputs Scenario 4 of all joint costs, 381,314 50,725 432,039 16.2 marginal cost of freight No=e(1) Costsco clculate avoidable cDst are as follows C(Qp.O)(only pseger) = 2.060.970billion yen C(O,Qf)(only fiecighr) = 381,314 billion yen Ct (passnger& fhtight)= 3,227.275billion yen Cq=C(Qp.QO(joint cost) = 784,991billion yen

TableV-19: Companison between Fae Leveland Costs JR pasen- JR freight ger revenue AC MC average rate fromtrack use (average cost) (mainal cost) Ton-kmbasis 7.29 0.65 43.7 1.9 Note: AC (averag cost) and MC (marginalcost) areestimated by the time-seriesresults(medtd1). 98 Japan( . National RailwaysPrivatization Study II Re.erences Caves,D. W. and L R. Christensen(1980), Mizutani, F. (1993), 'A Private-PublicCom- "Thc RelaciveEfficiency of Publicand parisonof Labor Produccivityand Utiliza- Private Firms in a CompetitivcEnviron- tion in JapanescUrban Railways,"in ment: The Case oFCanadianRailroads," Proceedingsof the Third International Joumal of PoliticalEconomy, Vol.88, Conferenceon Competitionand Owner- pp.958-976. schipin SurfiacePassenger Transport, Caves,D. W. and L R. Christensen,and JA Toronto, Canada (forthcoming). Swanson(1981), 'ProductivityGrowth, Mizutani, F. and L Nakamura (1994)," ScaleEconomies, and CapacityUdlization in EconomicAnalysis of ProductivityofJR U.S.Railroads, 1955-1974," Amcrican Eco- Railways,"Joumal of Public Utility Eco- nomicReview, 71,December, pp.99 4 10 02 . nomics,"Vol. 46, No. 2, pp. 121-145.The Fukui, K (1992), "Japani - National Rail- Japan Societyof Pubt;.:Utility Economics. waysPrivatizacion Sturdy; jApanese Expei- Moriya,H. (1991), "Investmentof Private encr and Lessonsfor DevelopingCountries," Railwaysin MetropolitanArcas," Econom- World Bank DiscussionPapers 172. ics Today, The Japan DevelopmentBank Gomez-Ibanez,J. A. andJ. R. Mcyer(1990), (japancse). "Privatizingand DeregulatingLocal Public Moriya,H. (1992), "RailroadFares, Profit- Services:lessons from Britain'sBuses," abilityand Capital Investment,"Economics joumal ofAmeriain PlanningAssociation, Today, The Japan DcvelopmentBank VooL56,No.1, pp.9 -21 . (Japanese). Gomez-Ibanez,J. A. and J. R. Mcyer(1993), Nakamura,K(1984), 'A Surveyon Productiv- Going PrivateThe InternationalExperi- ity Measurementin RegulatedIndustries," ence with Transport Privatization,The Journal of Transport Studies,The Japan BrookiingsInstitution, Washington, D.C. Societyof Tranport Economics,pp. 1 09- Meyer,J. R. and J. A. Gomez-lbanez(1980), 136 (japanecs). "Measurementand Analysisof Productvity Nakamua, K(1994),"Priyaizon andDereguLa- in Transportation Industr:s," in J. W. tionofNanl Monopoly:Cases in Railway Kendrickand B. N. Vaccara(ods),New R nmas",Wascd CommnialReview, No. Dcvelopmentin ProductivityMeasurement 359,WasedaUniveity (fapansc). andAnalysis,The Universityof Chicago ShleiferA.(1985),"A thcory of yardsticcom- Prcss,Chicago, IL, pp.295-332 . petition," Rand Journal of Economics, Meycr,J. R., J. F. Kain and M. Wohl (1965), Vol.16, No.3, Autumn, pp.319-32 7. The Urban TransportationProblem, Vickers,J. and Yarrow,G.(1988), Privatiza- Harvard Univ. Press,Cambridge, Mass. tion: An EconomicAnalysis, The MIT Miyajima,M. and S. D. Lce (1985), "Man- Press,Cambridge, MA. agement and Efficiencyin LocalPublic and Viton, PA.(1980), "On the EcAnomicsof Private Enterprises:A Comparisonin the Rapid-TransitOperation," Transportation Caseof Railways,"Joumal of Public Utility Research-A,Vol. 14A,pp.2 4 7-2 53. Economics,Vol.36, No.3, pp.6 9 -10 3 Yamaka,K and Ayuha, S. (1994), "Measures (apanesc). to Achieve Efficient Physica Distribution Mizutani, F. (1994), JapaneseUrban Rail- Systemin Japan", ResearchReport 189, ways;A Private-PublicComparison, The japan DevelopmentBank (Japanese). Avebury,Ashgate Publishing Company, Aldcrshoc,England.