Report No. 34220-IN India Road Transport Service Efficiency Study November 1, 2005 Energy & Infrastructure Operations Division South Asia Regional Office Document of the World Bank CURRENCY EQUIVALENTS Currency Unit = IndianRupees (Rs.) U S D 1 = Rs.43 Rs. 1 = USD 0.0232 1Crore (Cr.) = 10,000,000 ABBREVIATIONS AND ACRONYMS AIDS Acquired ImmuneDeficiency Syndrome AITD Asian Institute o f Transport Development A S R T U Association o f State Road Transport Undertakings BOT Build-Operate-Transfer CES Consulting Engineering Services (India) Pvt. Ltd. GDP Gross Domestic Product GO1 Government o f India GVW Gross Vehicle Weight HDM Highway Design and Maintenance System HIV Human Immunodeficiency Virus IRDA Insurance Regulatory andDevelopment Authority LCV Light Commercial Vehicle MAV Multi-Axle Vehicle MOC Ministryo f Communications M O S R T H Ministryo f Shipping, RoadTransport & Highways MVA Motor Vehicle Act MVD Motor Vehicles Department NHA National Highway Authority NRTC National Road Transport Commission OD Own Damage Rs. IndianRupees RTC Road Transport Corporation RTO Road Transport Office SOE State Owned Enterprise STU State Transport Undertaking T A C Tariff Advisory Committee TCI Transport Corporation o f India TIR Transport International Routier TPL ThirdParty Liability UNECE UnitedNations Economic Commission for Europe Vice President: Praful C. Pate1 CountryDirector: Michael F. Carter Sector Manager: Guang Z. Chen Team Leaders: George Tharakan, Zhi Liu ACKNOWLEDGEMENTS This study reviews the long-distance road transport industry in India in order to identify inefficiencies that could reduce the benefits to be derived from the large investments now being made by the Government in the nation's highway infrastructure. The findings o f this study have been presented at recent seminars such as the "Global Infrastructure Summit" organized bythe Federationo fthe IndianChambers o fCommerce and Industry (FICCI) inMarch 2005, and the "Convention on Reforms inthe Road Transport Sector" organized by the All India Confederation o f Goods Vehicle Owners Associations (ACOGOA) and the All India Bus Operators Confederation (AIPOC) in February 2005. It draws heavily on a number o f studies that has been carried out by Clell Harral, Ian Jenkin, John Terry, Richard Sharp, Eugene Gurenko, the firm Consulting Engineering Services, Inc.(CES), and the Asian Institute of Transport Development (AITD). It also relies on two complementary studies carried out for China by Jianfei Zhang, currently Director General o f the Ministry o f Communications China and for Pakistan by Sardar M.HumayunKhan, for internationalcomparisons. All these studies were commissioned andmanaged by the Energy and Infrastructure Unit (SASEI) o f the South Asia Region o f the World Bank for the purpose o f undertaking the subject sector work. The final report has beenprepared by a Bank team, including Alok Bansal, Isabel Chatterton, EugeneGurenko, Simon Thomas, Zhi Liu(Task Leader for early part o fthe study) and George Tharakan (Task Leader). N.S. Srinivas and RajeshSingh provided administrative assistance. The report has beenprepared under the general guidance o f Guang Z. Chen, Sector Manager for Transport, South Asia Region. Peer reviewerswere Paul Amos, Asif Faiz andGraham Smith. FOROFFIcI[AEUSEONLY This document has a restricted distribution and may be usedby recipients only in the performance of their official duties. I t s contents may not be otherwise disclosed without World Bank authorization. TABLE OF CONTENTS EXECUTIVE SUMMARY. FINDINGSAND RECOMMENDATIONS ....................................................... 1 ITHETRUCKINGINDUSTRY . ...................................................................................................................... 7 1.1STRUCTURE OFTHE HIGHWAY FREIGHTINDUSTRY ...................................................................................... 8 1.2 FREIGHTRATES AND INDUSTRYPROFITABILITY ......................................................................................... 10 1.3 QUALITY OF SERVICE ................................................................................................................................ 13 1.4 THEVEHICLEFLEET...................................................... ...................................... 14 1.5 COSTS OF TRUCKDELAYS FACILITATION AND PAYMENTS ......................................................................... 17 1.6 TRUCKING SAFETY.. ....................................... 22 1.7 AXLELOAD CONTROLS............................................................................................................................... 27 I1INTER-CITYBUSSERVICES . ................................................................................................................... 31 11.1 THEROLEOF THE STUS PAST. PRESENTAND FUTURE ............................................................................. 32 11.2 PRIVATE INTER-CITY B u s TRANSPORT ...................................................................................................... 39 11.3 REVAMPINGGOVERNMENT POLICY TO BETTER SERVE THE PUBLICINTEREST.......................................... 4 1 11.4 RECOMMENDED PUBLIC POLICY FORINTER-CITY BusSERVICES ............................................................. 44 I11COMMERCIALMOTORINSURANCE . ................................................................................................. 47 111.1CURRENTSTATEOFTHEINDIANC~MMERC~AL MOTOR INSURANCEMARKET ......................................... 47 111.2 STRUCTURALIMPEDIMENTSTO AN EFFICIENTINDIANAUTOINSURANCEMARKET ................................ 49 111.3 POLICY RECOMMENDATIONS .................................................................................................................. 50 I V KEY RECOMMENDATIONS AND THEIR IMPLEMENTATION . ..................................................... 54 ANNEX 1: ECONOMICOPERATING COSTSFORDIFFERENT TRUCKS ININDIA ........................ 56 ANNEX 2: COSTSDUE TO ADMINISTRATIVE CHECKSAND FACILITATION PAYMENTS .......60 ANNEX 3: PROPOSEDAMENDMENTS TO THE MOTOR VEHICLE ACT 1988 TO OVERCOME OVERLOADING ............................................................................................................................................... 66 ANNEX 4: CHINA. EFFICIENCYOF ROADTRANSPORTINDUSTRY THE ..................................... 68 ANNEX 5: PAKISTAN. ROADTRANSPORTINDUSTRY THE ............................................................. 73 EXECUTIVE SUMMARY, FINDINGS AND RECOMMENDATIONS 1. The World Bank has longbeeninvolved infinancing road infrastructure inIndia, and presentlythe Bank's program exceeds USD 4 billioninthe road sector alone. Government policies for the organization and functioning o f the roadtransport industry will have major implications for the economic returns and societal impacts o f these large investments inhighway infrastructurenow underway. This study has been undertakento assess the present policy regime, and identify measures which may be considered to improve the functioning o froad transport, in particular long-distance roadtransport, and enhance its already enormous contribution (3.9% o f GDP) to the workings o fthe Indianeconomy. While the roadtransport sector encompassesawide variety o f activities, this study has focused on three aspects which were considered the most relevant to the investments inhighway infrastructure-the trucking industry, inter-city buses, and inview ofits very importantbutlargelyunfulfilledroleinenhancingroad safety, the motor insurance industry. The key findings and recommendations o f the study are summarized below. Summary of Key Findings The Trucking Industry 2. Despite many remaining impediments, mainly concerningthe existinginfrastructure, India has achieved a highly competitive. low-cost road freight transport industryfor basic services, with highway freight rates among the lowest inthe world. The industryis deregulated and, as in manycountries, highly fragmented with many small operators. The industry's structure, comprising transporters, broker agents and small operators, i s market driven and appears to be serving the market reasonably well. Giventhe very low freight rates, one has to conclude it i s an effective industry structure. 3. Itis the constant pressure o fahighlycompetitive marketthat delivers to India's shippers some of the world's lowest freight rates. Insuch a competitive market, one would expect freight rates to vary little from costs, and our estimates o f costs confirm that expectation. Infact, trucking freight rates are so low that the industryis suffering an intense period o f low profits or even losses. - Inthis context, actions bygovernment that increase costs or reducethe efficiency ofoperators, will soon find their way into higher freight rates. 4. While the industry delivers very low freight rates, service quality is poor, with low reliability and transit times nearly double that o f developed countries. This low quality o f service maybe adequate for much o f the present traffic comprising low-value bulkproducts -much o f which would normally be served more economically byrailway or coastal shipping. However, it i s not adequate for higher-value manufactures or the time-sensitive export trade which comprise a growing share o fthe Indian economy. 5. Equipmentutilization rates for the Indiantrucking fleet, which average 60,000 kmto 100.000 kmpertruck-year, are less than a quarter o fthose indeveloped economies. These low utilization rates are caused by long delays at checkpoints enroute, excess trucking capacity which results inidle trucks, slow speeds on most roads, especially incongested areas, and lack o ftractor trailer unitsthat enable the tractor to keep operatingwhile loading and unloading are carried out on the trailers. - 1 - 6. Truck delays at checkpoints have been estimated to cost the economy anywhere between Rs.9 billion and Rs.23 billion a year in lost truck operating hours. Giventhe present surplus capacity inthe trucking fleet, the opportunity cost of the lost time would be less, but this estimate does indicate the magnitude o fthe problem o f truck delays. The estimate does not include "Facilitation Payments" made at the checkpoints to circumvent various regulations, and these have been estimated to range betweenRs.9 and Rs.72 billion. While these unofficial payments, being transfers, are not directly a loss to the economy, they probably result inrevenue and other losses, to the Government andthe economy, far inexcess o f the monies actually paid. 7. The trucking industrytoday usesmainly 2- and 3-axle rigid trucks with a small sleeper cab and an open top freight box o f 30 to 40 cubic meters. Given the competitive market conditions, it can be inferredthat the existing fleet mix i s overall the most economical giventhe array o f vehicles currently available to the Indian trucking industry. However, that i s likely to change as the roadnetwork i s improved, the mix o f traffic changes, andthe array o f available vehicles i s widened. Because oftheir low cubic capacity, current Indianvehicles are not so low cost when moving light-loading freight for which the freight rate i s almost doubled per ton-km due to the smaller weight o f cargo that can be accommodated. 8. An increase o f 10 percent inthe market share o f tractor-trailer units has been estimated to result ina reduction intransport costs on the order o fRs.5 billionper year. With improved fleet management enabling more intensiveuse o fthe tractor units, these potential savings could increase to Rs.8 billion. These units also have the advantage o f more modern technologies which enhance driver comfort and the safety o foperations. Introduction o ftractor-trailer multi-axle vehicles would reduce not only transport costs but also road damage caused by the higher axle-loadings o f 2- and 3-axle rigidtrucks. 9. Government needs to develop incentives to expand the multi-axle truck fleet as these trucks cause less pavement damage and are o fmore modern design resultinginlower per unit costs, higher fuel efficiency and reduced emissions o fpollutants. The incentivesproposed for introduction o fmulti-axle trucks include reduced tax andhighway toll rates inrecognitiono f the lower costs these trucks impose on public infrastructure. 10. India's legal single axle load limit i s now 10.2 tons, however, most Indianhighways, both national and state, were constructed for axle loads o f 8.16 tons, the previous legal limit. It has been estimated that strengtheningthis older network for the increased load limit would require investmentso fRs.200 to Rs.300 billion. Controlling axle loads is critical to protecting these investmentsonce they are made. To protect the investments inthe Golden Quadrilateral and its diagonals, it i s estimated that the physical infrastructure (weighbridges, etc.) for axle load controls would cost around Rs.2.5 billion, which i s well worth the expense considering the size o f investmentprotected. 11. The present single axle load limit o f 10.2 tons appears to be slightly lower than the optimal, which was assessedto be inthe range o f 11 to 13 tons. However, since a small degree o f overloading, say 5 percent, would be treated leniently, operators could approach the optimal when carrying highdensity loads without excessive repercussions. Hencethe current axle-load limit maybe considered to be sufficiently close to optimality. 12. Road Safety i s a major concern for India, with fatality rates about ten times those inthe developed economies, and trucks are responsible for a disproportionate share o fthese accidents. - 2 - The annual economic loss from road accidents has been estimated to exceed Rs.550 billion, with a majority attributable to the truck fleet. To improve the safety record, driver training, licensing, the working conditions of drivers, and enforcement o f safety regulations mustbecome a priority for the Government. Inter-City Bus Services 13. The Road Transport Corporations Act o f 1950, initiatedthe creation o f State Transport Undertakings (STUs) with monopoly franchises for many inter-city bus services. At their zenith in the early 1980s, the STUs controlled 45 percent o f India's bus fleet. While they incurred losses on most urban services, their rural and inter-city services generally covered operating costs. The Motor Vehicle Act of 1988 reversedthe policy, and encouraged greater reliance on the private sector, which led to increased competition for the STUs and mounting losses. Bv the year 2000, STU losses exceeded Rs.22 billion; the Government then halted financial support to the STUs and encouraged State Governments to do the same. The role o f STUs ininter-city bus services i s receding inmost o f India, and several states now rely exclusively on private sector provision. 14. Despite restrictive granting o fpermits and unfavorable/discriminatorytax treatment for private operations, the private sector has won back a rapidly increasing share o fthe inter-city road passenger market, and now about 80 percent o f the bus fleet i s privately operated. A study o f private inter-city bus operations inthree states has confirmed that there is a significant amount o f clandestine operations by private operators who provide stage carriage services while holding contract carriage permits. However, the services appear to be generally satisfactory, with between 60 and 70 percent o f users surveyedrating the services either satisfactory or good. 15. Unitcosts ofSTUoperations have escalated due to excessive staffing, andonaverage STU costs per passenger-kilometer are more than 40 percent higher than that o fprivate operators. STUstaff costs are now about three times that o fthe private sector; they employ on average 7 staff per bus at an average salary o fRs.7,700 permonth, whereas for private operators those numbers are 4.3 and Rs.3,500 respectively. Reducing STU staffing and salaries to levels comparable with the private sector would result inan annual savings o f around Rs.40billion. Not all of this would accrue to the economy, since some o f it i s a transfer from STU staff to bus passengers or tax payers. Redeployment o f surplus STU staff, however, would be a true saving. 16. The Association of State RoadTransport Undertakings (ASRTU) has made aproposal for reform, the thrust of which generally supports corporatization o f STUs, subsidies, compensation for social mandates imposed by Government, cross-subsidies among routes, andan expanded role for STUs inthe regulation of inter-city services. This report's recommendations, however, differ from the ASRTU proposals insome important respects : it i s doubtful that the social obligations imposed on STUs, which the ASRTU would like to see paid as a subsidy to the STUs, could not be delivered more efficiently by the private sector; the case for the capital and route operating subsidies advocated by the ASRTU inthe inter-city bus markets i s at best weak; their proposed bundlingo funprofitableroutes with moreprofitable ones inorder to provide a cross-subsidy to the former has generally been found unworkable; and the proposal that STUs act as bothoperator and regulator presents serious issues o f conflict o f interest andmoral hazard. 17. Iftheprimaryobjectiveisthebest quality service at the lowestpossible cost, the long-run strategy for inter-city bus services must be to move the STUs to majority private ownership in competitive markets as early as possible. Arguably, competition is more important than - 3 - privatization. However, so long as there are large labor forces employed by publicly owned carriers with highunit costs, the pressures to restrict competition and keep tariffs hiphto protect the least efficient carrier will remain strong, and this will protect not only the STUs but also serve as an umbrella to protect marginally efficient firms inthe private sector as well -all at the expense o f higher fares for bus users or higher taxes for the citizens at large. 18. more than 700,000 -andtheir present unprofitability, privatization cannot be accomplished Giventhe magnitude o fthe labor engaged by the STUs -their total labor force numbers overnight but will require a phasedprogram. A central element o fthis prowam must be labor force reduction, retraining and re-deployment. This can be accomplished over timebynatural attrition by not replacing employees as they retire or leave; by offering early retirementincentives to accelerate the process; byjob retraining (either for outplacement or to fill different jobs within the STU); and by offering improvedtermination grants. Such measures can cushion the impact on affected employees and also provide well equipped labor to meet the broader economy's needs. 19. The appropriate focus o fregulatory policy inthe case o froadpassenger transport should be qualitative standardsrelatedto the safety o f services, and the minimization ofnegative environmental impacts. Safety dimensions encompass vehicle roadworthiness standards (brakes, steering, tires, visibility, lightingand signaling), driver qualifications andworking hours, and avoidance o f overloading. With regardto the environment, standards should bephased into require low emissionbuses, and also control the disposal o f lubricant wastes and other materials. Motor Insurance Industry 20. Motor insurance accounts for about 40 percent o f the gross insurance premiums written inthe Indiannon-lifemarket, or over Rs.60billion, whichmakes itthe largest lineofnon-life insurance business inIndiatoday. Most o f that i s written by the 4 public sector companies. While no accurate statistics are available on how muchi s from the coverage o f commercial vehicles, according to some estimates, commercial policies account for about 50 percent o fthe market. Insurance companies must use a standardinsurance policy form issuedby the Insurance Regulatory and Development Authority (IRDA) which covers both Own Damage (OD) and Third Party Liability (TPL). Tariffs are regulated. TPL i s compulsory but compliance i s poor with an estimated 50 percent o fvehicles not covered. 21. The motor insurance industryinIndia is not performing- a critical roleneeded to enhance road safety penalizing poor driver performance through increasedpremiumsor denial of cover. - For liability insurance inIndia it i s the vehicle, not the owner or driver, which i s insured. Thus it i s the vehicle's accident record that impacts on the experience rating aspects o fthe insurance premium. Consequently, anowner or driver with abad accident record canreplace the vehicle and thus avoid an adverse experience-rated premiumincrease. 22. Auto and truck liability insurance policies do not have an upper liability limit, while premiums are generally controlled, and as a result, loss ratios on truck insurance exceed 100%by a wide margin. Providing liabilityinsurance is consequently not a business that the insurance industrypursuesor seeks to develop. Removing tariff controls and allowing a free market to develop will enable the industrvto turn this into a viable business and to invest inthe kinds o f enhancements needed, e.g. a system to maintain and access driver records inorder to properly assess risk and charge premiumsthat reflect the risk profile o f individual drivers. - 4 - 23. There i s a strongneed for creating an integrated insurance claims database that can be shared iointly bv the Indian insurance market. Such a databasewould prevent adverse selection and the possibility o fbad drivers taking advantage o f information asymmetry. More selective underwritingmay, however, leave some bad driverswithout insurance cover, and for them a motor insurance pool with punitive tariffs could be established. 24. The legal framework for the motor insurance industry, under the Motor Vehicle Act o f 1988. lacks features common inmore advanced legal frameworks such as a statute o f limitations, liability limits and thresholds for claims adiudication. These are critical to the efficient and effective finctioning o fthe system, and their absenceresults ina variety o f abuses and shortcomings such as fraud, uncertainty for reserving against future claims, and excessive numbers o f claims resulting ina backlog inthe motor tribunal system. Summary ofMainRecommendations Regarding the Trucking Industry: 25. To reduce delays at border crossings, particularly for highvalue or time-sensitive goods, thereportrecommends considerationo fa systemsuch as the EuropeanT.I.R., to permitsealed trucks which elect to use the system to operate without en-route inspections on the basis o f a certificate issued at origin by a duly authorized and bonded issuing entity. 26. To encourage use o f multi-axle vehicles and tractor-trailer combinations, thereby reducingtransport costs and roadpavement damage, it i s recommendedthat incentives beput in place such as tax rates favoring such vehicles and reduced tolls on highways to reward their reduced impact on pavements. 27. Since a significant portion o f the driver population i s illiterate, it i s recommended that audio-visual driver training materials be developed. 28. To prevent excessive hours o f driving, it is recommended that trucks operating outside their home stateberequiredto carry two licensed driversat alltimes. 29. To improve axle load controls, changes recommended are: expand enforcement authority beyond officials o fthe Motor Vehicles Department; distinguish between minor (up to 5% o f gross vehicle weight) and more excessive overloading for which there would be extreme penalties; and make abetment an offence to enable action against the broker or transporter arranging the load. 30. Invest inpermanent weigh stations at strategic locations on the National Highway network to enable random checks o ftrucks passing the weigh station when the station i s open. Require trucks found to be over-loaded to unload the excess load at their own cost andrisk. Regarding Inter-City Bus Services: 31. The STU reform proposals advanced by the ASRTU are unlikelyto produce the desired improvements ininter-city bus services or stem the losses incurred by the STUs, and consequently, it i s recommendedthat the strategy for STUreforms be reviewed. - 5 - 32. Reforms inthe Inter-City Bus services sector should include deregulation o f tariffs, restructuring and commercialization o f STUs, elimination o f STU monopoly rights, changes in the tax regime to achieve uniformity o f tax treatment o f all buses operating inthe inter-city markets, and creation o f an independent agency to establish, monitor and enforce competition rules and ensure access to common user infrastructure (terminals, bus stops). 33. The appropriate public policy for the inter-city bus services sector would be to remove quantitative regulations restricting entry into the inter-city bus transport markets, and to allow market forces to determine bothtariffs and the types o f services offered. Regarding the Motor Insurance Industry: 34. Itis recommendedthat switchingto a system where experience-rated premiums attachto the owner and the driver, not to the vehicle, be taken up as a matter o fhighpriorityby IRDA. 35. IRDA should initiatethe development o fanintegratedclaims database. 36. IRDA should also explorethe creation o fa motor insurance pool for baddrivers who have been denied cover by the insurance industry. 37. Finally, the report recommends amendment o f the Motor Vehicle Act o f 1988 to remedy deficiencies with respect to motor insurance such as the lack o f provisions regarding a statute o f limitations, liability limits and thresholds for claims adjudication. - 6 - EFFICIENCY OF ROAD TRANSPORT SERVICESININDIA 1.1. The World Bank has long beeninvolved infinancingroad infrastructureinIndia and presently the Bank's transport program exceeds USD4 billion inthe road sector alone. Government policies that govern the organization and functioning o fthe road transport industrywill havemajor implications for the economic returns and societal impacts o fthese large investments inhighway infrastructurenow underway. This studyhas beenundertaken to assess the present policy regime, and identify measures which may be considered to improve the functioning o f road transport, inparticular long distance road transport, and enhance its already enormous contribution (3.9% o f GDP) to the workings o fthe Indian economy. While the road transport sector encompassesa wide variety o f activities, this study has focused on three aspects which were considered the most relevant to the investments in highway infrastructure -the trucking industry, inter-city buses, and inview o f its very important but largely unfulfilledrole inroad safety, the motor insurance industry. The recommendations hereinare consistent with the 2004 Country Assistance Strategy and the India PolicyReview document "India: The Challenges Ahead" prepared by the Bank in2002. 1.2. The report is presented inthree sections corresponding to each o f the three main topics: the trucking industry (Section I), inter-city bus services (Section II), commercial and motor insurance (Section 111). The three sections maybe read as stand-alone pieces, and readers interested in aparticular topic may go directly to the relevant section. The main conclusions and recommendations o fthe report are presented inthe combined Executive Summary for the benefito freaders desiring a quick overview. I.THETRUCKING INDUSTRY 1.3. Despite many remaining impediments, mainly concerning the existing infrastructure', Indiahas achieved a highly competitive, low-cost road freight transport industry for basic services, with highway freight rates among the lowest inthe world (see Table 1.1) and indeed surprisinglylow giventhe operatingconditions inIndia. The development o ftrucking industrypolicy inIndia inrecent years generally follows the recommendations ofthe 1999 Sundar Committee report entitled "Trucking Operations inIndia".2 Presentations by trucking association officials at a workshop initiating the present study on January 8,2003, and subsequent interviewswith trucking operators and state government officials3have generally echoed the picture painted by the Sundar Committee. Officials o fthe Ministry o f Shipping, RoadTransport and Highways (MOSRTH) have confirmed that the Sundar Committee Report i s still a primary guidingforce inthe Government's trucking industrypolicy. The large scale improvement o f road infrastructure through the National Highway Development Program, the Improved Riding Quality Program, and the increased assistanceto the State Governments for improvement o f road infrastructure are important measures already implemented by the Central government to address infrastructure deficiencies. AsianInstitute ofTransport Development for the Ministryof Surface Transport, TruckingOperations in India: Report of Steering Committee (November 1999, hereinafter cited Sundar Committee). The study team visited India January 5-28,2003 and conducted field observations of trucking operations and interviews with trucking operators, association representatives, central and state government officials inNew Delhi,Maharashtra, Karnataka, and Uttar Pradesh. - 7 - Country Average cost per ton km (US$) Pakistan 0.015-0.021 India 0.019-0.027 Brazil 0.025-0.048 United States 0.025-0.050 Central Asian republics 0.035-0.085 Australia 0.036 China 0.040-0.060 SOURCE:Clell Harral,Ian Jenkins,John Terry, RichardSharp, TheEfficiency ofRoad Transport in India: The Trucking Industry, WB BackgroundPaper, 2003. 1.4. Although this note generally echoes the Sunder Committee recommendations, it differs on two important aspects: (i) degree o f government involvement through the regulation; and (ii) use o f the trucking industry as a mechanism to enforce varying state sales taxes. As well demonstrated byworldwide e~perience,~the trucking industry normally performs best where government involvement i s limitedprimarily to provision o f good road infrastructure, enforcement o f safety aspects, and internalization o f external diseconomies, i.e. the costs to society caused but not directly paid bythe roaduser (e.g. pavement damage from too heavy axles, congestion, and environmental effects). Further government interventionhas the potential to do more harm than good. 1.5. The MotorVehicle Act o f 1988 made major advances inthe liberalization and harmonization o f government regulations across the different states o f India, but there remains much to be done to improve and streamline administration, reduce corruption, and enhance enforcement. Reliance on the trucking industryinIndia as an instrumentfor enforcement ofvaryng state sales taxes across the statesplaces an additional heavy administrativeburdenon the sector, and it i s not clear whether the introduction o f the Value AddedTax will ease that burden. 1.1Structure of the HighwayFreightIndustry 1.6. India's trucking industry i s deregulated and, as inmany countries, highly fragmented.5 It is primarily composed o f three types o f enterprises as described below. Percentage of operators with Percentage of vehicles in fleets of size Less than 10 trucks Over 100 trucks Less than 10 trucks Over 100 trucks Great Britain 94 0.2 51 12 USA 79 2 19 45 Japan 42 2 11 19 1.7. Transporters: These are trucking companies which have the primary contact with shippers and receiving customers. They solicit fi-eight, largely on an annual price quote basis, bill, collect, and carry the accounts receivable, are responsible for cargo loss and damage claims and perform the other customer service functions. Some, like the Transport Corporation of India (TCI), are fairly substantial enterprises with manybusiness locations. They typically own a fleet of trucks and often warehouses andterminals as well. But they relyprimarily on small truck operators for their line-haul (intercity) transportation. Inthe United States they might be called broker-operators, although the latter would, inaddition to cargo insurance, also carry third party liability insurance covering their trucking operators and sub-contractors. 1.8. Truck Operators: These individuals (often called owner-operators) typically own one or a very small fleet o f trucks, which usually are financedby high-leverage debt. They usually drive the main truck and the other drivers are family members or are personally known to the owner. InIndiavirtually all o fthese trucks are two driver units usually with a helper. These people virtually live inthe truck, following the traffic flows and goinghome primarily when the load they are carrying takes them nearby. They contract with the transporter to perform the intercity transport o f freight which the latter has contracted for with the shipper. The truck operator performs almost all the freight transport activity on a shipment and i s paid either an agreed percent o fthe revenue or a flat amount. It appears the amounts and percentages paid the operators fluctuate continuously reflecting the supply- demand situation inthe individual commodity and origin-destination segment o f the market. 1.9. Brokers,Agents: These parties haverelations with the truck operators andprovide those operators to the transporters. They play a necessary role inIndia because they act as a quality control on the reliability o f the truck operator and a means o f facilitating prompt loadingby the operators. With limitedmeans o f assessinga truck operator's record o f timely, claim-free performance, the transporters, who have the customer relationships and are responsible for cargo loss and damage, need some way o f determining the reliability of the truck operator. The brokers or agents take on that job. Inthe current Indian freight market, with an excess of trucks, they also seek out loads for their following o ftruck operators. They are usually paid a fixed amount per load; 300-500 Rupees was mentioned as the rate range in January 2003. 1.10. This industrystructure i s market driven and appears to be servingthe market reasonably well. Customers need a substantial partyto be their transporter because o f the needfor cargo loss protection, a stable freight price and a reliable supply o fhauling capacity. The labor laws, road delays, and predominance o f live-loading make small truck operators a lower cost alternative to big, company-owned fleets, so the transporters primarily use the small truck operators insteadof their own trucks. Inthis context, there is a real needfor a facilitating middleman, the broker, and thus the market has created that activity. All three groups are working ina market-determinedrevenue sharing cooperation. This is the way the deregulated Indian freight market has organized itself. Given the very low freight rates, one has to conclude it i s an effective industry structure. SOURCE: Jenkins IA. "All change -new directions for the road transport industries of Russia, Ukraine, Kazakhstan and Belarus", TransportReviews, 1994, Vol. 14, No. 4,289-320. - 9 - 1.2 FreightRates and Industry Profitability 1.11. Freightrates vary with the interplay o fcommodity type, size o fshipment and the volume freight to bemoved between any pair o fpoints on any given day inrelationto the supply o fhauling capacity available. Transporters have been forced by the much larger customers to quote annual rates at which they will move goods. They assert these rates are adhered to, though there i s no way to prove that statement. However, they admit that the amount they have to pay to the truck operators on any given day can vary from half o f their revenue to more that 100% o fthe revenue. 1.12. The Economic Times and other newspapers publishfreight rates daily; those inthe Economic Times are for outbound Delhi9 ton full loads.6 It i s highlyunlikely that these rates are lower than actual market rates, as shipper representatives would be very reluctant to tender freight at a rate above the publishedones; there is evidence to suggest that actual rates are sometimes discounted from the publishedrates. Publishedlong distance road freight rates inmid2002 were about Rs.10.50 per truck km for a standard nineton capacity truck, which is equivalent to Rs.1.17 per ton km assuming on average a full load, but not an overload (see Table 1.2 for details). These rates are not regulated by government and are considered to reflect market conditions. It i s reportedthat rates can rise by up to 50% during the autumnpeak agricultural seasonand fall by 20% or so at slack times, but the midsummer rate i s close to the annual average. Due to imbalances infreight demand, freight rates vary with direction on some routes. 1.13. Inpractice the trucks often carry more than the registered capacity. Inearly 2003 truckers were reporting rates o f about Rs.9-11 per km for trucks carrying normal loads, but Rs.13per kmifoverloaded by20% or more above the registered capacity. This could imply that freight rates per ton km are even lower than shown inTable 1.2. 1.14. The publishedrates were inline with the rate quotes givento the mission during its field work, and there is reasonable confidence that the publishedrates are representative o f the average charges for legal, i.e. not overloaded, trucks. Rs.10.50 per truck kilometer, or Rs.1.17 (US$ 0.023) perton km, has beentaken to represent areasonable overall average. These freight tariffs are among the lowest inthe world, as noted earlier inTable 1.1, which shows rates from several countries with competitive trucking markets. Table 1.2 PublishedLongDistance RoadFreightRatesinIndia SOURCE: Mid2002 ratesfor 9-tontrucks, advertisedinEconomic Times of India,FinancialExpressof India;TCI The Sundar CommitteeReport (p.199)includesa table ofmarketbasedMumbaioutboundfreightrates inJune-October 1998. The currentnewspaperrates appearhigherthanthose o f the Sundarreport,butbyamountsthat seemreasonablein relationto the time gap betweenthe dates. - 10- 1.15. These freight rates are so low that one can readily accept the assertions o f all those interviewed that the industryi s sufferingan intense period o f low profits or losses.' Insuch a competitive market, one would expect freight rates to vary little from costs, and our estimates o f costs confirm that expectation. Pakistan's trucking industryi s facing a similar situation. 1.16. One component-by-component estimate o fthe average financial operating cost o f trucks in India i s shown inTable 1.3, as provided in2002 by the Transport Corporation o f India (TCI), which is the largest operator inthe country. These figures include administrative overhead costs that small operators would not normally incur, but do not include interest charges or any profit margin. While any transporter may have incentives to overstate its operating costs, the costs shownby T C I for the nineton capacity truck (Rs.10.46 per truck km)hasbeencross-checked against HDMmodel estimates' and agree closely also with India's competitive market freight rates described above for this type o ftruck. 1.17. We have estimated the financial operating costs for small operators, which make up the majority o fthe industry, as shown inTable 1.4. Consistent with the mission interviews with Indiantruckers inthree states inJanuary 2003, these estimates assume lower utilization, lower staff earnings and less overheads, but include interest payments and a greater allowance for broker payments. Ifa modest profit margin o f 4% o f other costs i s included, the total costs per truck km for the dominant %ton truck operated by the small operator would be Rs.10.77 (US$ 0.215) per truck km; excluding any profit, the total i s Rs.10.38 (US$20.8) per truck km. Allowing a reasonable marginfor error, these numbers are essentially the same as those reportedby TCI. 1.18. These estimated costs, inrelation to the low tariffs, lend support to the assumption o f only a very modest profit level or break-even situation for the industry as a whole currently, with frequent bankruptcies at the margin. Ifthe tariffs are now below the longterm variable cost, they mustultimately adjust to that level, as the operators and transporters are all small, thinly-capitalizedbusinesses, usually with a highdegree o f financial leverage. Revenue levels below variable cost will soon put enough capacity out o fbusiness that freight rates will rise to the level needed to draw investment back into the industry. Inthis context, actions by the government or market forces that increase costs or reduce efficiency o f the operators, especially inareas of equipment utilization, soon find their way into higher freight rates. 1.19 The transporter, such as TCI, also operates on very thinmargins. For its 10-15% o f the revenue it must solicit, bill and collect for the freight, and bear the cost o fbad debts and cargo claims. It must also carry the investment indriver advances, usually 50% to 70% o f the amount ultimately due the driver, andthe slow accounts receivablewhich appear to average about 60 days. The transporter also has to quote a relatively fixed annual freight rate which leaves it financially exposed. The transporters are probably reasonably profitable, but it is unlikelythat they are generally able to take advantage o fthe truck operators, for there is also a competitive market amongtransporters insearch o f truck operators and truck operators openly compete against the transporters wherever possible. Similarly, the pure agentbrokers face competitive markets for their services and would not generally be able to take advantage o f the truck operators. '*The truckers' strike that occurredApril 13-23 hrther underscores this point. IanJenkins, "India: Efficiency o fthe Road Transport Industq- Modeling andAnalysis" (HWTSL Informal Technical Note, May 18,2003) - 11- NOTES: (1) Assuming an average price of Rs.ZO/litre for fuel. (2) Depreciationbased on resale after five years (45% of purchase price) and truck prices of Rs.600,000 (5 ton), Rs.800,000 (9 ton), Rs.1,050,000 (16 ton) and Rs.2,250,000 (27 ton). No interest charges are included. (3) Other costs includebrokerage and vehicle insurance. SOURCE:TCI (mid 2002), as reportedby Davis(2002) --- Profit Depreciation 60,000 80,000 105,000 225,000 Other 27,900 54,300 72,700 112,400 15,900 31,100 41,500 64,200 Total Overheads 148,290 235,490 297,930 519,510 Total Cost 441,840 861,790 1,152,330 1,781,610 Annual Utilization (km) 45,000 80,000 80,000 80,000 Cost per truck km 9.82 10.77 14.40 22.27 Cost per ton kmof capacity 1.96 1.20 0.90 0.82 - 12- 1.20. Enterprises similar infunction to the transporter are common inthe U.S., and they generally average about 15% o fthe revenue for their activity. One expects that these U.S. counterparts are much more profitable because o f the almost complete absence o f cargo loss exposure, the faster collections and almost complete lack o f credit losses, and the fact they do not have to guarantee the freight rate to their customer. This tends to confirm the view that the transporters arenot taking advantage o fthe truck operators inIndia. It is the highly competitive nature o f the marketthat makes the truck operators' business so hard, at the same time delivers to India's shipperssome o fthe world's lowest freight rates. 1.3 Quality of Service 1.21 Accurately measuringtransport service i s a statistically demandingtask because o f the many aspects o f service (including speed, frequency, reliability, and security), the ebb and flow o f freight volumes, and the many origin-destination pairs o fpoints. The mission could not attempt it, and was not able to locate any systematic survey o f transport users, i.e. shippers, on the subject. But discussions with the transporters, examination o fthe terminal facilities, and observations from intercity roadtravel done by the mission leads to the opinion that a relatively poor quality o f service i s rendered. The service may be adequate for much o f the present traffic comprising low-value bulk products- much ofwhich would normally be served more economically by railway or coastal shipping. It is not adequate for the higher- value manufactures or the time-sensitive export trade that comprises a growing share o fthe Indianeconomy. Incomparison, China's andPakistan's (Annexes 4 and 5) highways are better and there are much fewer inspections enroute, which results infaster service. 1.22 There was an excess o ftrucks everywhere the mission travelled (Haryana, Maharashtra, Karnataka, and Uttar Pradesh), so timely pick up seems assured. But the over- the-road speed looked very slow. The government-imposedmultiple check point system, the mixed use o f the roads by motorized and non-motorized traffic, the almost complete lack o f highway safety enforcement, and the poor quality o f the equipment has made transit times slow and reliability o f timely delivery i s almost certainly very poor bymodern standards. 1.23. For example, truckers estimated that a Delhi-Mumbai load (1408 kilometers) takes three days; a Delhi-Bangalore load (2,019 kilometers) four to five days. Inbothcases the truck would have one or two drivers (excluding helper). The U.S. equivalent transit times, with one driver operating legallywould betwo and three days (actually second and third morning), respectively. Ifthe US.carrier used2-driver teams, a day would be cut from each movement. 1.24. One positive service aspect noted was that the loads are generally properly tarped to protect against weather damage. This suggests that the economic linkage between the transporter, who pays the cargo claims, andthe truck operator, who is the personprotecting the loadfrom weather damage, isworking at least to some degree. - 1 3 - 1.4 The Vehicle Fleet 1.25. The Indian industry today uses mainly 2- and 3-axle rigid trucks with a small sleeper cab and an open top freight box o f 30 to 40 cubic meters (1100 to 1400 cubic feet). The low cubic capacity reflects the present freight market o fpredominantly heavy, often unpackaged commodities. The trucks are o f old, low-tech design with 135 to 165 horsepower naturally- aspirated engines. Tata and Ashok Leyland designs predominate with very limited numbers o fthe large new Volvo tractor-trailer units ina few market segments. Interms o ftraffic on the main national highway network, the 2-axle 9-ton truck constitutes roughly 75 percent o f the trucks, as shown inTable 1.5. The Indian vehicles are low cost trucks; new unit costs quoted are US$16,000 for the 9-ton andUS$20,000 for the 16-tontrucks, but the purchase price o fthe higher powered and modernized Volvo i s quoted at US$ 84,000 to 90,000, a relativeprice difference that has undoubtedly contributed to Volvo's limitedmarket penetration so far. Compared to India, Pakistan has a larger proportion o f 3-axle rigid trucks (9%) and multi-axle trucks (25%) inits fleet. InChina, the proportion o f medium trucks i s declining andthat o f light vehicles and tractor-trailer units are increasing. Table 1.5 EstimatedAverage Traffic Flow on the Main National Highways (2002) NOTE: Typical average traffic composition on the Golden Quadrilateral and main East-West andNorth-South Roads. SOURCE: Consultant's estimate based onnationaltraffic statistics, Davis (2002), and classified counts described inRoad Maintenance and Corridor Management for National Highways System inIndia (2000), Feasibility Report Consultancy Package VI for World Bank National Highways Project (1999), and CRRI, Updation o f Road User Cost Data (2001). 1.26. Giventhe competitive market conditions intrucking, it canbeinferredthat the existing fleet mix i s overall the most economical from the array o fvehicles currently available to the Indian trucking industryunder present conditions o f the road network and mix o ftraffic. The Indiantruckers appear satisfiedwith their trucks and generally runthem for their entire lifetime -as long as 20 years. However, that i s likely to change as the road network i s improved, the mix o ftraffic changes, and the array o f available vehicles i s widened. Such factors are likely to play a growing role inIndia, as it transforms from a - 14- largely rural agrarian economy to an increasingly urban, export-oriented manufacturing economy with rapidly growing high-tech and services elements. 1.27. Because o f their low cubic capacity, current Indianvehicles are not so low cost when moving light-loading freight. The effect on transport costs can be seen by comparing light- and heavy-loading freight. InIndia cube i s converted to weight for billingpurposes at 12-15 kilograms per cubic foot. So the 1100 cubic foot freight box can carry 13,200 kilograms, clearly an overloaded truck, as commonly observed. But the average kilos/cubic-foot inthe current U S road freight market, for instance, i s inthe range o f 4-5. So the 1100 cubic foot box could only handleabout 5,000 kilosbefore cubing out. Yet it would still cost 10.50 rupees per kilometer to move those goods, and the cost per ton kmwould be at least 1.8(9,000/5,000=1 .S) times as high as today's more typical cargo. 1.28. The contrast between the freight transport needs o f Karnataka and other states i s interesting. Inother states traffic such as bagged grain, cement, and steel rods dominate the traffic flow, but inKarnataka computers and other hightech products are muchmore prominent. The open top rigid trucks that commandthe Indianfreight market, with their low line-haul speeds and unreliability, are hardly suitable for the just-in-time logistic needs o fthe computer industryand other high-tech undertakings. Indiawill need to introduce highcube vans and fast line-haul transit inthe near fbture to serve these needs. Prospective Savings from Wider Adoption o f Tractor-trailers 1.29. Introducingmore tractor-trailer multiaxle trucks into the Indian truck fleet would affect the overall cost o f transport, including: the operating costs o f the vehicles; the quality of trucking services; roadprovision costs; and environmental and other externalities. An attempt has been made to assess the cost impacts o f introducing multi-axle tractor-trailer unitsinincreasing proportions into the vehicle fleet. 1.30. This analysis i s concerned only with illustrating the potential differences invehicle operating costs. A number o fhypothetical scenarios are defined: 0 first, for different mixes o f vehicle sizes, inwhich the road freight market share o fmultiaxle tractor-trailers varies from 10% at present up to 40%, and 0 second, for different ratios o ftrailers to tractors for multiaxle vehicles. The first o fthese sets o f scenarios can indicate the potential cost savings o f introducing more tractor-trailers into India, assuming that no increase inutilization and efficiency o fthe tractor trailer operation takes place; the benefit under this scenario i s solely through the substitution o f larger vehicles for smaller vehicles. The second set o f scenarios indicates the potential additional cost savings that can be achieved ifefficiency o ftractor-trailer operations i s increased by more intensive scheduling and increasing the ratio o f trailers to tractors, so that any single tractor can service several trailers and thereby reduce its waiting time for loading, thereby increasing the productivity o fthe most costly component inthe system 1.31 The analyses inthis case are interms o f economic rather thanfinancial costs, with taxes excluded from financial costs to derive a simple estimate o f economic costs inkeeping - 15- with current planningpractice inIndia. Results o fthe analyses are summarized inTable 1.6 and discussed below. Increasing the Market Share o f Tractor-trailers 1.32 As indicated inTable 1.5, tractor-trailers currently constitute only about 2% o fall motorizedtraffic on the main highways o f India (the Golden Quadrilateral and main east- west and north-south roads). This is only about 4% o ftruck km. However allowing for the different carrying capacities o fthe trucks, the current market share o ftractor-trailers on these roads i s 10% as shown inTable 1.6. The impact o f increasingthe market share o f tractor- trailers has been assessedfor three possible alternative cases: Case A: Increasingthe market share from 10% to 20%, Case B: Increasingthe market share from 10% to 30%, Case C: Increasingthe market share from 10% to 40%. 1.33. The corresponding vehicle kmon the main highways are shown inTable 1.6 -this raises from 1,016 million at present up to 3,928 million for Case C. Indefiningthese alternative scenarios it was assumed that the light truck market share would remain unchanged because this type o ftruck serves its own market niche. For simplicity and to avoid introducing effects other than that o f increasing the fleet o f tractor-trailers, the market share o f 3-axle trucks i s also assumed to remain unchanged -implyingthat the number o f 3-axle trucks that operators replaced with tractor-trailers would be balanced by the number o f 2-axle trucks that operators replaced with 3-axle trucks. 1.34. The annual economic vehicle operating costs o fthese alternative scenarios are also shown inTable 1.6, assuming the vehicle operating costs for each truck type estimated earlier. It i s assumed that the increase intractor-trailers only occurs on the mainhighways and not on other national highways. Inpractice tractor-trailer operation could also be extendedto most o fthe nationalhighways, so this underestimates the potential impact. 1.35 According to these estimates, the annual vehicle operating costs o ftrucks on the main highways inIndiacould be reduced from about Rs.207 billion at present, to as little as Rs.190 billion for Case C. The estimated reductions are between Rs.5 and 6 billion for Case A, almost Rs.12billion for Case B and over Rs.17billion for Case C. Increasing the TrailedTractor Ratio 1.36. The impact of increasingthe ratio oftrailers is also indicated inTable 1.6for boththe Cases A and C, assumingthe vehicle operatingcosts vary for differentratios as showninTable Al.4 of Annex 1. For Case A, the potential reduction intotal costs o ftractor trailers usingthe main highways i s about Rs.2.8 billion, from Rs.201.4 to 198.7 billion, assuming that the ratio o f trailers to tractors can be increased up to 2.0. For Case C the potential reductioni s estimated at around Rs.5.6 billion. - 16- Table 1.6 Road Freight Costs with IncreasingTractor-Trailer Use -5 100% 1 - 210,000 90% J 80%- - 205,000 "3z 7 0 % - II) - 200,000 60% - 50% - - 195,000 40% - - 190,000 -Ye Heavy 3-Axle - 185.000-%Light Vehicle 10% X Tractor-Treilarl Multi-AxIm 0YO L - 180,000-% Heavy 2-AxIe Present CaseA Case B CaseC +Economic C O S ~ S Note: CaseA Increasing the Tractor-Trailer market share from 10%to 20% CaseB: Increasingthe Tractor-Trailer market share from 10%to 30% Case C: Increasing the Tractor-Trailer market share from 10%to 40% 1.37. Itshouldbenotedthatthepotentialbenefits ofincreasing the trailer/tractor ratio would be even greater for the modern units now beingintroduced into India. Becauseo fthe higher capital cost ofthese units, higher utilizationisnecessaryto exploit hllythe potential advantages o fthese units interms o flower operating costs and/or greater performance and reliability. Howeverat present it i s not possible to estimatethe relative costs andbenefits of usingthe modemunits due to lack ofinformationabout operatingcostsunder Indian conditions. 1.5 Costs of Truck Delays and Facilitation Payments 1.38. The annualkilometer runbyintercitytrucks varies between60,000 kmand 100,000 km; and80,000 kmhasbeentaken as areasonableestimatefor the averagelongdistance unit. Almost allunits are usingone or two drivers andmost also have ahelper, atotal oftwo or three people inthe cab. As an illustration o fgood utilization, two driver units inthe U.S. (called sleeper teams) would average about 400,000 kilometers per year. 1.39. The low equipment utilizationhas anumber ofcontributing causes: (a) Unliketractor-trailer equipment, which dominatesinter-city servicesinthe UnitedStates andother countries with well developed roadtransport systems, the rigidequipment that dominates the Indiantruck fleet requires powerunit downtime while the truck is loaded or waiting for a load. The analysis inSection 3 above illustrates the potentialthat might berealized from widescale useo ftractors with multipletrailers. (b) The governmentcheckpoint systems consumeline-haul hours inunproductive waiting. Truckers reported that 15%-25% o f line-haul time was lost at checkpoints, although other estimateswere as low as 4%. The stridency oftruck operator - 17- complaints, and observations from time to time o f long queues o ftrucks along the highways, confirm this as a realproblem, although it appears to vary considerably from location to location andtime to time. An estimateofthe likelyrange ofthe downtime dueto this problemispresentedbelow. (c) There is currently an excess o ftrucks inthe Indianmarket; the mission observed large numbersofemptytrucks sittingidle duringtimes trucks would normallybe loading and was consistently toldby all parties interviewedthat there is a large surplus oftrucks. (d) Trucking operators are reluctant to operatetrucks without a substantially full load, and are often preparedto wait untilthe traffic materializes. divided four lane roads -compels low speeds. (e) Thetraffic mix o fnon-motorized andlow-powered vehicles -even on (f) Theunder-powered andoverloadedtrucks themselves arenot capable offast highway speeds, evenon uncongestedroads. Loaded truck speeds of only 40 kilometershour are typical. The numerous old trucks onthe highways are, in addition, delayedby frequent breakdowns. (g) Inadequacieso fthe existing highwayinfrastructure, includinglack of city by- passroads, also lower averagespeeds andannualkilometers traveled. Outside the limitedmainarteriesofthe nationalhighwaynetwork, roadandtraffic conditions typically do not allow sustainedhighspeed operations. Truck Delays Dueto Administrative Controls 1.40. Inaddition to the delays causedbynormaloperational factors such as traffic congestion andthe need for driver rest andvehicle maintenance, long distancetrucks inIndia experience significant delays due to governmental administrative activities, such as control of sales tax collection. These delays can occur either at fixed check points or at road-side locations set up bymobile enforcement units. Delays even occur before the trip begins, while the driver completes the necessarytaxation proceduresfor carrying the load. 1.41. The adverse impacts onproductivity of conflicting administrative rules o fdiverse jurisdictions are difficult to measure, but the cross-country comparisons o fregulatory reforms conducted for this studyg-as well as the quantification attempted below -suggest that these penalties are severe. Stateand localregulations are often enactedto serve parochial interests that do not contribute to national transport productivity. It is important to note that improvements inroadtransport performance inthe UnitedStates after economic deregulation R. Sharp et al., "A CrossCountry ComparisonofRegulatory Reforms to Promote RoadTransport Efficiency," (HWTSLInformal Technical Note, May 2003). - 18- in 1980were onlypartly attributable to federal deregulation itself. A host o fother changes occurred to harmonize state regulations ranging from licensing, to vehicle size and weight standards, to driver and vehicle safety measures. InAustralia, evidence o fthe value o f such harmonization is even more clear. While economic regulations hadbeen abolished for years, substantial productivity gains inthe trucking industry were achieved after a mechanismwas created in 1992to achieve greater standardization and performance orientation instate motor carrier safety, vehicle and traffic control standards, licensingandinsurance. 1.42. International experience does not sustain the argument that all state rules for motor carriers needto be made uniform or be replaced by a single national standard. Indeed, the Australian experience is notable inthat the commonwealth has very weak regulatory powers and the states have harmonized their procedures with the facilitative assistanceo f a committee structure with no enforcement capabilities -the National Road Transport Commission (NRTC), which i s somewhat similar to India's Transport Development Council. Important Australian reforms havebeen achieved byNRTC promoting a common vision, supported byresearch that demonstrated the efficiencies to be expected from state adoption of less restrictive performance standards and other liberalization measures. Whether such a voluntary cooperative mechanismo f the various states inIndiacould be equally effective i s a matter that will best be assessedbythe respective Indian authorities. Truck Delay Costs 1.43. The economic cost ofdelays to trucks due to administrative controls hasbeen estimated from: 0 the amount o ftraffic on the nationalhighways (truck kmper year), 0 the extent o f traffic delays at check points or at the start o f the trip (hours per truck kmor per trip), and 0 the unit cost o f the delays (Rs. per truck hour) in terms o f vehicle operating costs and cargo time costs. Economic impact has been assessedpurely interms o f user costs. There are other costs, including the administrative controls themselves (labor and other administrative costs) that are not included here. 1.44. Unfortunately, reliable information about delays and their causes is lacking. The evidence from available formal surveys" indicate more limited delays than those reportedby transporters and truckers interviewedby the study team. A simple approach i s therefore adopted inorder to indicate the approximate level o f delay caused by all the various causes. Since there is considerable uncertainty about the delays, lower andupper estimates o f the economic impact are given to indicate the sensitivity o fthe result to alternative assumptions. 1.45. Estimates of the annual economic costs to Indiao f administrative delays to trucks are given inTable 1.7. These estimates include costs o f delays on the whole national highway network but exclude costs o f delays on state and other local roads where the loThe only formal surveys that have come to the attention o fthe study team are the three done bythe Central Institute ofRoadTransport inJune-July 1998 for the Sundar Committee and reported inAnnexure 3.1 o fthe Sundar Committee Report, although other less formal trip reports have also beenreviewed. - 19- likelihood o f delay due to administrative control i s relatively small. A detailed discussion o f the derivation o fthese two tables is given inAnnex 2. 1.46. From Table 1.7 it can be seenthat truck delay costs due to governmental administration formalities are estimated to lie somewhere within the rather broad range o f Rs.9 billion to Rs.23 billion, reflecting the fragmentary and uncertainnature o f the available information. However, even the lower limit o f this range, Rs.9 billion CUSS180million) per annum, which may be considered a quite conservative estimate, i s a significant magnitude, and the true numbermight easily be twice that. "Facilitation Pavments" l1 1.47. It is also a common practice for roadtransport companies and drivers to make (unofficial) payments to govemment officials to expedite clearances andor to avoid payment o fheavier taxes or fines. For example, vehicle overloading i s a very common practice, and truckers report that as much as 10% o f their freight revenues may be taken up inpayments to facilitate passage of overloaded vehicles; some states have defacto legalized this practice by the issuance o f formal permits, but the issue i s probably more often dealt with byillicit payments. Available information about such "facilitation payments" made by truck drivers in the course o fjourneys i s analyzed to assess the impact o fthese payments on transport costs. These are real financial costs to the truckers (which are undoubtedlypassed through to shippers, given the structure o fthe industry), but they are merely transfer payments (or income redistribution), and do not constitute economic costs (additional consumption o f economic resources). 1.48. The estimated range of "facilitation payments" i s evenwider, as mightbe expected consideringtheir mostly illicit nature, runningfrom as "little" as Rs.9 billion to as highas Rs.72 billion, as shown inTable 1.8. Inthis case, however, it must be recognized that these payments are often made to avoid larger, typically much larger, payments legally due, whether excise or other sales taxes or fines for violations o fthe law. The loss o frevenues to the govemments concerned is most likely a multipleo fthe estimates given inTable 1.8. Those cases involving failure to enforce traffic safety laws contribute to even larger economic and social losses from accidents that mightbe avoided. ''Inthe United States the term "facilitation payment" i s often usednarrowly to indicate a payment to get an official to expedite an action that he is legally boundto do inany case; here it is used more broadly to encompass also bribes to get an official not to take an action that he i s legally boundto take. - 20 - Table 1.7 Estimated Annual EconomicCost of Administrative Truck DelaysinIndia -- higherestimate lower 895 6,576 1,074 887 9,433 estimate 2,188 16,076 2,626 2,169 23,059 NOTE: (1) Traffic on mainhighways (13,252 kmconsistingof the Golden Quadrilateralandmain east-west andnorth- southroads) is estimated as showninTable 1.6. Traffic on otherhighways(44,448 kmofnationalhighways) is assumedto behalfthose onthe mainhighwaysand delaysper kmare also halfthat on mainhighways.(2) The same user cost per hour is assumedfor all highwaysandthis is consistent with the finding of the CRRI surveys which indicatedthat the average value of commodities doesnot vary greatlywith category ofhighway. SOURCE:Consultant's estimate (A) Paymentson MainHighways - Vehicle km(million) Low Estimate High Estimate 27,087 27,087 - Paymentper km(Rs.) 0.13 1.oo Total User Payment (Rs. million) 3,521 27,087 (B) Paymentson Other Highways -- Vehicle km(million) 45,426 45,426 Paymentper km(Rs.) 0.13 1.oo Total User Payment (Rs.million) 5,905 45,426 (C)TOTAL, USERPAYMENTS (Rs.million) 9,426 72,513 NOTE: Basedon equivalent assumptionsto those made inTable 7. SOURCE: Consultant's estimate - 21 - 1.49. Table 1.9 summarizes the costs o f various inefficiencies inthe trucking sector, and shows that between Rs. 17 and 46 billion o f economic costs could be saved per year should the inefficiencies inthe current systembe addressed. Table 1.9 EstimatedRangesofEconomicCosts Savedas aResult of Reforms Area of Refordmpact Vehicle Fleet Improvement Notes: (1) These are real financial costs to the truckers (which are undoubtedlypassedthrough to shippers, given the structure of the industry), but they are merely transfer payments (or income redistribution), and do not constitute economic costs (additional consumption o f economic resources). (2) Assuming that the market share o f tractor- trailers increases to 20 percent. (3) Assuming that the market share o f tractor-trailers increases to 40 percent. (4)Assuming that the ratio o f trailers to tractors can be increasedup to 2.0. and that the market share o f tractor- trailers increases to 20 percent. (5) Assuming that the ratio of trailers to tractors can be increased upto 2.0 and that the market share of tractor-trailers increases to 40 percent. Source: Compiled by the World Bank using Clell Harral, Ian Jenkins, John Terry, Richard Sharp, The Efficiency o f Road Transport inIndia: The Trucking Industry, WB BackgroundPaper, 2003 1SO. A possible solution is to adopt a system like the EuropeanTIR Camet developed by the UNECEto facilitate cross-border movement o f goods. The system i s described insome detail inthe box overleaf. While the TIR was developed for use by vehicles crossing nationalborders, there i s no reason why the concept could not be adapted to serve the needs of inter-state goods movements inIndia. The control issues facing various State authorities inIndiaalong a truck's inter-state itinerary are similarto those facedbynationalauthorities at borders, andthe costs and delays which the TIR was intended to eliminate are the same as the ones now faced by Indiantrucks at these border crossings. It shouldbe noted, however, that such a systemwould mainlybe suitable for the movement o fhighvalue goods inclosed trucks, which are the very cargoes most affected by frequent inspections and consequent delays at checkpoints. Shippers o f such highvalue goods are also more likely to be willing to pay for the costs o f the system. These costs would include the administrative costs o fthe system and the costs o f underwritingthe guarantees for the payment o fvarious duties and fees applicable along the itineraryo fthe truck. 1.6 Trucking Safety 1.51. While road accident statistics are notoriously flawed and often conflicting, road accidents are unquestionably a grave problem inIndia-with a magnitude dwarfing the HIV/AIDSproblem -andone which will getworse as levels o fmotorization rise from their -22- present very low levels. Best estimates by specialists place highway deaths in India at about 75,000 per year with an annual economic loss (excluding any allowance for pain and suffering) o f approximately Rs.550billion (US$l 1billion).'2 Trucks are reportedly involved inabout halfofall highway accidents, and a disproportionate numberare causedbya relatively few accident-prone drivers. Inthe majority o f cases (65%), rash and negligent drivingis identified as the cause ofthe accident. However, it appears that those who cause accidents -whether rare or fiequent offenders -very commonly escape the consequences. Driver training, licensing and labor conditions inthe trucking industry, as well as strong deterrents to unsafe drivingthrough the operation o fthe insurance system, are critical issues for the improvement o f trucking safety. THETIR SYSTEMROLEIN FACILITATINGGOODS MOVEMENTS BORDERS ACROSS The TIR Convention has proved to be one o f the most effective international instruments prepared under the auspices o f the United Nations Economic Commission for Europe (UNECE). To date, it has 64 Contracting Parties, including the European Community. It covers the whole o f Europe and reaches out to North Africa and the Near and Middle East. Already today, the United States o f America and Canada are Contracting Parties as well as Chile and UruguayinSouth America Traditionally when goods crossed the territory o f one or more States in the course o f an international movement o f goods by road, the Customs authorities in each State applied national controls and procedures. These varied from State to State, but frequently involved the inspection o f the load at each national frontier and the imposition o f national security requirements (guarantee, bond, deposit of duty, etc.) to cover the potential duties and taxes at risk while the goods were intransit through each territory. These measures, applied ineach country, ledto considerable expenses, delays and interferences with international transport. Inorder to ensure that goods maytravel with aminimuminterference ''en route" andyet offer maximum safeguards to Customs administrations, the TIR regime contains five basic requirements -the five pillars o fthe TIR Customs transit system: Goods should travel inCustoms secure vehicles or containers; Throughout thejourney, duties and taxes at risk should be covered by an internationally valid guarantee; Goods should be accompanied by an internationally accepted Customs document (TIR Carnet), opened inthe countryo fdeparture and serving as a control document inthe counties o f departure, transit and destination; Customs control measures taken inthe country o f departure should be accepted by all countries o ftransit and destination; Access to the TIR procedure for o national associations to issue TIR Carnets; and o natural and legal persons to utilize TIR Carnets shall be authorized by competent national authorities. The successof the TIR systemmaybejudgedby the numbero f Carnets issued each year. As a result of the expanding East-West European trade, particularly since 1989, and the corresponding increase in international road transport, the number o f TIR Carnets issued exceeded one million in 1992and bv the end o fthe 1990shadreached around three million. - 23 - Enforcement o fRegulations 1.52. It is paramount that the Government takes actionto enhance enforcement o f existing regulations and laws -including road safety aspects, drivers licensing, axle load limits, and vehicle permitting-with the objective to reduce unsafe driving, improve vehicle utilization, andexpedite cargo movement. Enforcemento fvarious provisions o fMotor Vehicles Act and Central Motor Vehicles Rules i s often fixstrated by use o f fake drivinglicenses or Registration Certificates by many vehicle owneddrivers. This problem can be addressedto some extent bythe computerization o fRTO offices. Improvedrecords keepingwith hll scale computerization to allow cross checking, is one important step. Some progress has beenmade concerningthe control o f fraudulent documentation. Standardized software for bothback-end and front-end operations has been developed andmade available free o f cost to all the States, byMORTH. More than 10percent o fthe existingRTOs have already done back-end computerization andmany other RTOs are inthe process o f doing so. Soon, Smart Card basedDrivingLicenses and Registration Certificates will be issued with interoperability o f the systemthroughout the country. TruckinP Labor 1.53. It is estimated that the trucking industry employs over 4 million drivers inIndia, and this numberis likelyto increase rapidly with the expansion o fthe highway network, andthe consequent increased role o f trucking infreight movements. However, unlike most other transport sectors (aviation, the railways, inland water transport or shipping), the conditions o f labor, its training and licensing, and the penalties for negligent conduct are the least developed inthe trucking sector. Some key information on the driver population and their conditions, based on a survey conducted in 1998, i s presentedinTable 1.10. Literate 27.33 % Non-Matriculate 29.04 % 5 to 8 hours 28.88 % 9 to 12hours 44.59 % More than 12hours 20.39 % InsuranceCover Carried Only Third Party Liability About 20 % ComprehensiveandTPL About 80 % - 24 - 1.54. The data inTable 1.10have important implications for the development o f the trucking industry and the Government's policies inthe sector. With 20% o f truck drivers illiterate, clearly driver training programs cannot rely solely on written materials for the training and licensing oftruck drivers. There i s a need to develop audio-visual training and testing materials inthe local languages to serve the illiterate portion o f the truck driver population. The survey data also indicates the importance o f affordable arrangements for driverrest and relaxation. This shouldbe animportant consideration inhighway facilities design. Safe andbetter controlled environments inrest areas, while improving driver performance, could also reduce the incidence o f sexually transmitted diseases for which long distance truck drivers are second only to commercial sex workers, as the most important disease vector. That almost halfo f the drivers surveyed admitted to drivingbetween 9 and 12 hours a day is also a cause for concern, andregulationsrequiring two drivers inthe cab at all times may be the only way to ensure that a tired driver does not feel compelled to continue drivingto meet a delivery schedule. Insurance Issues inTrucking Safety 1.55. As clearly demonstrated incountry after country, an effective insurance industry has a critical role to play inimproving the safety o fthe roadtransport system. However, for reasons that are not entirely clear, the Indianinsurance system has never assumed the pivotal role inhighway safety that it has inother c~untries.'~Supporting law enforcement in imposing accountabilities for unsafe drivingis, o f course, not the only function o fthe insurance industry, but arguably it i s its most important one from the perspective o fpublic p01icy.'~ The problems relatedto the insurance system are described inthe following: 1.56. Inliabilityinsurance inIndiait isthevehicle, not the owner or driver, whichis insured. Thus it is the vehicle's accident recordthat impacts on the experience rating aspects o fthe insurance premium. Since most accidents are due to the performance o f either the driver o f the vehicle or the owner who controls the driver, owners and drivers should bear the economic brunt of experience-rated premium increases ifthe drivingor accident record is not good. As it stands now, it i s a simple matter for an owner or driver with a bad accident recordto replace the vehicle andthus avoid an adverse experience-rated premiumincrease. Experience-rated premiums should attach to the owner and to the driver o f the vehicle. 1.57. Inauto and truck liability insurance the policies do not have an upper liabilitylimit while premium rates are controlled. The result has been loss ratios on truck insurance that exceed 100% by a wide margin,I5 obviously decreasing any incentives the insurers may have to pursue such business. This is due inlarge part to government regulatory controls on premiums charged. Insurance industryrepresentatives indicate that political lobbying by trucking interests has kept the premiums down. This constitutes an implicit subsidy to l3 Sundar Committee Report, Chapters 9 and 13. Much of the ensuing discussion draws heavily o n these two chapters of the Sundar Committee Report, plus an interview with Shri KK Bhat (who was the member o f the Sundar Committee from the insurance industry) and several interviews with truclung industryrepresentatives in January 2003. l4 Road Safety Guidelinesfor the Asian and Pacific Region, Asian Development Bank, (1998). l5 The Sundar Committee (pp 75-76) reported that insurance claims outgo has o n average been around 350% o f premium income for Third Party Liability coverage, but only around 55% for the carrier's OwnDamage coverage. - 25 - trucking, but one mainly realized by unsafe vehicle operators. Consideration should be given to allowing the insurance camers greater flexibility to set policy limits and deductibles and to have more freedom inadjusting insurance rates on an owner or driver experience-rated basis. Some additional flexibility has been granted recently, whereby insurers can increase liability rates up to 400 percent of the base rate for a driver with a poor driving record, however, they cannot refuse to issue liability insurance. 1.58. Accident investigation andrecords-keeping are inadequate, and the insurance industry have either not beensufficiently motivated or empowered to compel improvements. Police are supposed to collect records relating to accidents, but according to the Sundar Committee, the Motor Vehicle Incident (MVI) reports mostly contain a statement that records were not produced, so it is not possible for the insurance company to prove any violation.16 1.59. The driver licensingsystem is badly broken. It i s commonly stated that a majority o f truck drivers have invalid commercial driver licenses. It i s imperative for highway safety that the systemo f driver licensing function properly. This must include not only proper and reasonable driver qualification and testing, but also an open system so that insurers, vehicle owners and others with valid needs can get complete information on an individual driver's accident and violation record. This information will help insurers and truck operators get unsafe drivers off the road.17 1.60. Inadditionto accident liabilities, there are also problems inthe cargo insurance area. The transporter is liable for cargo loss or damage though inmost cases the preventiono fthat loss and damage is inthe hands o fthe truck operator. While it i s possible for the transporter to obtain carrier's liability coverage for the cargo, the insurance i s high, and apparently i s only available as part of a broader policy that i s difficult or impractical for small transporters to obtain. The owner of the goods can obtain cargo coverage from insurers and usually does get such coverage on loss- or damage-sensitive cargo. Inthe event o f a cargo loss the insurer pays the claim and then proceeds via subrogation to collect its loss from the transporters. This activity is only effective against the more financially established transporters; the small truck operators usually escape liability for their losses. Combined with the deficient driver licensing system, the ultimate result is bothbad economics andbad safety performance. In conjunctionwith the overall reforms inthis area, transporters shouldbe allowed to buy individual cargo coverage for the goods intheir (or their subcontractor's) possession on a basis that will draw insurance industry interest to that line o f coverage. 1.61. A reviewo f government policies that may inhibitIndia's recentlyprivatized insurance industryfrom playing a muchmore prominent role intrucking safety i s one o f the more important and urgent policy actions that needto be considered by the government. The l6Sundar Committee, p. 80. l7Itis should be noted that, prior to 1986, truck drivers inthe United States could obtain multiple licenses from different states and could conceal a poor safety record by keeping one license "clean" from any blemish. In 1986 a Federal statute was enacted which pre-empted the states' rights inthis area and required a national registry o f commercial vehicle operators inwhich they couldhave only one license, their "CDL." Even though the United States has had a shortage o f long distance truck drivers, drivers with bad violation or accident records now cannot readily findjobs. See R. Sharp et al., "A Cross Country Comparison o f Regulatory Reforms to Promote Road Transport Efficiency," Annex o n Truck Safety Regulation(HWTSL Informal Technical Note, May 2003). - 26 - Sundar Committee did not fully appreciate the contributionmade by the insurance industryto an orderly and safe trucking industryinother countries, and therefore did not lay sufficient stress on resolving the problems o fthe insurance industryinIndia. However, the Insurance Regulatory Development Authority (IRDA) has constituted a Committee headed by Shri Money whichwould be looking into these aspects. Inview of its importance to the effective, from a safety viewpoint, regulation of the auto industry, and inparticular its trucking sector, we have addressed issues inthe auto insurance industryinsome detail inChapter 3. 1.7 Axle Load Controls 1.62. As notedearlier, the country's surplus oftrucks andthe industry's ownership structure, with a majority of trucks owned by small operators owning one or two vehicles, makes the road freight industryinIndiahighlycompetitive. Consequently, the market operates at very low margins, with freight rates at times falling below operating costs. Given the pressures to turn a profit, andwith lax enforcement o f axle load controls, overloading of trucks i s a widespread problemwith axle loads often well inexcess o f the legal limit. Excessive axle loads have serious implications for the deterioration o froads, and for the funding requiredto keep roads maintainedat anacceptable level. 1.63. The legal axle load limit inIndiais 10.2 tons, as amended by the central government in1996under Section 58 ofthe MotorVehicles Act of 1988. However, most Indian highways, bothnational and state, were constructed for an axle load o f 8.16 tons, the legal limit beforethe 1996 amendment. Ithas beenestimated that strengthening this older network for the revisedlimit would require investments o f Rs.20,000 cr to Rs.30,000 cr. Controlling axle loads i s therefore critical to protecting these investmentsonce they are made. OptimumAxle LoadLimit 1.64. Theoptimum axle load limit is defined as the value o fthe axle load limit which minimizes the cost o f operatingthe vehicles, including the freight vehicles for a given tonnage o ftraffic, and constructing and maintaining the road network, such that the road condition does not deteriorate below an acceptable value of roughness. The damage caused by anoverloaded vehicle increases exponentially (to the fourth power) with anincrease in axle load, and consequently has a large impact on the condition o fthe roadpavement, which inturnaffects theoperatingcosts ofallvehicles onthe road. To assessthe optimal axle load under actual conditions, two road sections were analyzed based on detailed traffic androad condition data, which were used to simulate the effects o f different axle load limits. 1.65. The two corridors, Agra-Bharatpur-Jaipur (NH)andSirhind-Morinda-Ropar (SH), for which detailed data were available on pavement structure, traffic and axle-load distribution, were subjected to detailed studyusing the HDM-4 model to determine the optimum axle- load. The model was calibrated for each of the two cases to simulate the progression of pavement deterioration to reflect actual experience. The modelwas then usedto determine the optimum maintenancepolicy to minimizethe total cost for each case. The number o f vehicles were calculated for the total tonnage using the axle-load distribution subject to varying axle load limits. The numbers o f 2-axle trucks which carried less than 10tons were kept constant. Similarly all MAVwhich carried less than20 tons were kept constant. For all other vehicles the remainingtonnage was distributed equally among vehicles based on the definedaxle-load limit to determine the number of 2-axle andMAV inthe stream for the - 27 - given axle-load limit.The results o f the simulation for road agency costs, vehicle operating costs and total system cost for various axle-load limits are shown below for the two cases. Table 1.11: Agra-Bharatpur-Jaipur NationalHighway Agra-Bharatpur Section -a- R-BC -t-R-SDBC -STGI -STGZ -+-sTG~ +STG.I 3 :- 1500 I450 g2 ,- g11U)o -Ee,;I350 $1300 t- 1250 c m t- 0 1200 1150 I100 Axle Load (Tonnes) Bharatpur-Jaipur Section Table 1.12: Morinda-Ropar State Highway A R - B C --+-R-S[lBC t - STGI +STG2 d S T G 3 -STG4 4a 14 42 13 BAU Rile Lord $Tonnes) SOURCE: Consultant's estimate. - 28 - 1.66. From the above figures it appears that the optimumaxle-load limit ranges between 11 and 13 tons. A higher axle loadlimit, while reducingthe costs for vehicle operatorswould impose higher road agency costs for stronger pavements or increased maintenance expenditures. The present legal axle load limit o f 10.2 tons appears to do the opposite, i.e. increase vehicle operator cost (per ton-kilometer) andreduce road agency cost more than would be optimal. Nevertheless, since a small degree o f overloading, say 5 percent, would be treated leniently, operators could approach the optimal when necessary (highdensity loads) without excessive repercussions. Hencewe might conclude that the present legal axle load limit o f 10.2 tons is acceptably close to optimality. Legal Framework for Implementationo f Axle-load Limits 1.67. Overloading is an offence under the Motor Vehicle Act 1988, and is punishablewith fines. The statuary provisions dealing with overloading are set out inSections 113, 114, 194 and 200 o f the Act. These are briefly described below. 0 Section 113 limitsthe weight o f each laden vehicle to its registered gross vehicle weight. 0 Section 114 empowers the authorized officer o f the Motor Vehicle Department to enforce the weighingofthe vehicle andrequires the driver to unload the excess weight at his ownrisk and cost. 0 Section 194punishesthe violation o f Section 114 through fines based on the amount o f overloading. 0 Section 200 deals with compounding o f offences. 1.68. While these sections act as a deterrent to overloading, they do not have sufficient clarity and as a result different States interpret them differently. Many States at one time had issued a `golden' pass to the vehicle operators allowing them to overload by paying an additional fee. The Act does not providefor minor infringements nor does it provide severe punitive action for really egregious infringements such as a refusal to allow inspection. Another difficulty with the present act is that even though Section 114provides for unloading of the excess weight, the authority to enforce is confined to officials o fthe Motor Vehicle Department (MVD). Giventhe large area of a State andthe limited capacity of the State MVDs,it isnotpossible for the MVDsto enforce the act effectively. 1.69. Specific amendments to these sections o f the Act have beenproposed (see Annex 3) andthese are summarized below: 0 Expandenforcement authority beyond officials o fthe MVDs, to include other duly authorized agencies o fthe State, e.g. State police. 0 Distinguish between a minor overloading infringement (5%) versus more excessive overloading. 0 Make abetment o f overloading an offence, thereby enabling the punishment o f entities beyond the trucker, such as the brokeror transporter who organizes the load. 0 Introduce the possibility o f imprisonment for repeated offences. - 29 - Enforcement o f Axle-load Limits 1.70. Enforcement o f axle load controls, however, remains the most difficult challenge. This will requireboth investment inphysical facilities, and introductiono f appropriate procedures for inspectionand the imposition o fpenalties. The following are recommended for enforcement o fthe axle-load limits: 0 Involvethe truckers associations ininforming their members about the need to limit loading o ftheir vehicles to the legal load and secondly insupporting the government inits enforcement efforts. The association could also beempowered to assistthe operator unload and safely store any excess weight carried by vehicles detained at check points. 0 Governments should set up permanent weigh stations at strategic locations to enable random, surprise checks on vehicles passing the weigh station when checking i s underway. All vehicles found to be overloaded by more than 5% o f the GVW should be detained andrequiredto unload the excess weight before they are allowed to proceed on theirjourney. Sufficient space needs to be provided at the weighing stations to park overloaded vehicles for some duration. Facilities to unloadload the cargo, rest, etc. also needto be provided at these locations for a fee. 1.71. No detailed estimate o fthe needed investmentininfrastructurehas beenattempted here. However, at approximately Rs. 2 crores per weigh station andwith an average spacing between stations o f around 100 km, the investmentneeded to protect the Golden Quadrilateral and its East-West and North-South diagonals would be around Rs. 250 crores. Considering the size o f the investmentprotected this i s well worth the expense. Policy Changes to Encourage Use o fNew Technologies 1.72. As noted elsewhere inthis report, the Indiantruck fleet i s largely made up o f 2-axle rigidtrucks with extremely outdated technology. Introductiono fnewer technologies is necessary for a number o f reasons not least the need to reduce the excessive axle loads imposedby two-axle rigidtrucks. However, for technological upgradation to occur two conditions need to be met -increased competition inthe truck manufacturingindustry, and changes inthe cost structure to make the multi-axle truck more attractive to operators. 1.73. A set of incentives are needed to encourage the manufactureo fmodernmulti-axle trucks. This could be achieved through differential tax treatments which recognizethe lower costs imposedbyvehicles incorporatingfeatures such as multi-axles, air suspensions, improved tyres, etc. This would help improve vehicle performancenotjust on axle loadings, but also on safety and driver comfort andproductivity. Tax rates for such improved vehicles shouldbe fixed at lower rates to encourage their use. For example the road tax on an MAV could be the same as for the 2-axle truck, rather than a linear function of the load carrying capacity, as at present, with insome cases even an increasingunit rate per ton o f capacity. Similarly, road tolls couldjustifiably discriminate against two-axle trucks since they are the ones causing the most pavement damage. - 30 - 11.INTER-CITYBusSERVICES 2.1. Prior to the RoadTransport Corporation (RTC) Act of 1950, roadpassenger transport inIndiawas entirelyprovidedbyprivatelyownedoperators. However, the Government of India (GOI) was persuaded that the services on offer by the private sector up to that time were inadequate, and GO1inturnpersuaded the respective state governments to directly provide public transport for urban as well as inter-city and rural services. The State Transport Undertakings(STUs) were granted monopoly franchises on key segments o fthe market, and, drawing on capital from the states as well as the central government, they grew several folds over the next three decades. By the time they reached their zenith inthe early 1980s, they owned some 45 percent o f the bus fleet inIndia. Up untilthat time, while the STU urbanoperations incurred losses, the rural operations generally covered operating costs. Thereafter operations of the STUs were increasinglypoliticized, they were used for more and more employment generation, fare discounts were offered to an ever wideningbody o f special interest groups, and costs spiraled." 2.2. The Motor Vehicle Act o f 1988 reversedpolicy directions andencouraged greater reliance once again on private sector p r o v i ~ i o nby liberalizing marketentry inall market ' ~ segments except for certain `nationalized' stage routes where the STUs still retain some legal monopoly rights. The resurgence o fprivate operators providing an increased range of services, and inmany cases improved quality, incompetitionwith the STUs further aggravated the latter's financial problems. By 1991-1992 the STUs began to generate large deficits for rural as well as urban services, which reached more thanRs.2,200 crores (US$ 440 million) overall in 1999-2000. Pushingfor greater financial discipline and increasingly willing to rely onprivate sector solutions, the GO1has halted financial support to the STUs and encouraged state governments to do the same. The role o fthe STUs ininter-city transport is now receding throughout most o f India, and several states (mostly ineastern India) now rely exclusively on private provision. 2.3. To bridgethe gap between demand and supply, where the STUs have not met market demands, private operators have circumvented restrictive regulatorypolicies to operate clandestine services. Most o f the private buses having Contract Carriage permits violate their permit conditions andoperate as stage carriages. They compete with the multitude o f smaller vehicles such as maxicabs, jeeps, vans and LCVs andtourist cabs, all o f which operate freely without any regardto the permits they hold. These small vehicles do not adhere to any regulations concerning fares, routings or schedules, as their operations are fully flexible and can vary instantlywith traffic demand. As a result Indiahas a rather flexible inter-city passenger transport system, very responsive to the specific needs ofpassengers. It is unfortunate that this has been achieved at the expense o fpublic law, typically leaving the licensed franchisee with the burden o fthe unprofitable mandatorypublic services and the government with that o f enforcing the regulations. ''`* M.Koteeswaran, "Policy Issues Concerning RoadTransport" Associationof State RoadTransport Undertakings (undated, www.asrtu.org). See also M.Koteeswaran, "The State of the Public Transport" (paper resented to the Initiating Workshop for this study o n January 8, 2003). Consulting Engineering Services (India), "Privately Provided Inter-city Passenger Transport Services," Report to the World Bank (June 2002). - 31 - 2.4. Inthe next section we briefly examine the past andpresent role o fthe STUs, estimate the potentials for savings ifthe present legally monopolized STU `stage' services were to achieve the same unit costs as private operators presently do, and consider measuresto transition the STUs into a new structure to better meet hture societal needs. Inthe third section we assess the present inter-city bus services provided bythe private sector. The fourth and final section, considers the policy framework for inter-city bus services that could best serve the public interest. 11.1 The Roleof the STUs Past,PresentandFuture 2.5. The RTC Act o f 1950created the STUs "in the public interest and for public purpose.. . At that time, this Stateactivitywas not considered a business nor was profit a motive. Discharging social obligation providing basic facilities to the people at large was the prime consideration.. . The result was phenomenal. The villages that hadnever seen automobiles.. . saw buses on the roads to take them to places. At the time when the personalized motorized transport systemwas absent, the bus system integratedthe villages and towns, the rich and the poor, the rural and the urban."2o Since then, circumstances have swung strongly against the STUs. The very purpose o f the STUshas beentwisted to serve the political patronage objectives o fthe state governments, both as a source of employment generation andas a source o fpatronage by fare discounts to wider andwider segments o fthe population (students, senior citizens, the handicapped, medicalpatients, freedom fighters, language fighters, etc). These concessions have cost the STUs hundreds of crores that have neverbeen compensated bythe state governments (although some STUs are reportedto be compensated by government for these concessions). 2.6. Unitcosts ofthe STUshaveescalatedbecause ofhighlevels ofredundant employees- with seven employees perbus on average -andthe successful lobbying o fthe largelabor forces that obtained wages that are now no less thanthree times the levels o fthat o fprivate operators. The quality of service offered by the STUs has also failed to keep abreast o fmarket expectations. 2.7. Nonetheless, the STUs still remain the largest players inthe inter-city bus industry in most o f India. Inthe year 2000-2001 they provided no fewer than 10.7 billionnon-urbanbus kilometers carrying a total o f 375.1 billionpassenger kilometers (pkm) o f service. Continued reform o fthe STUs to better serve the public interest is thus a matter o f vital concern. 2.8. It is to the credit o fthe Association o f State Road Transport Undertakings (ASRTU) that much has been done to establish management information systems and make available to the public detailed information on the performance and costs o f each o fthe STUs so that performance o f each can be compared andjudged accordingly.21 These benchmarking *'M. Koteeswaran, "Policy Issues Concerning Road Transport" Association o f State Road Transport Undertakings (undated, www.asrtu.org). 21Profile and Performance, a thick compendium with detailed statistics on every S T U i s published periodically bythe ASRTU. Data for this report is drawn ftomASRTU, Profile and Performance 2000-2001. Itis not clear, however, to what extent the ASRTU independently audits the statistics reported by the various STUs to ensure accuracy or comparability. - 32 - systems not only help identify problem areas for management attention, they also create a kindof surrogate competition, which, though ratherweaker thancompetitionfor survival in the market place, can nonetheless stimulate efforts to improve. 2.9. Before turningto an agenda for STU reform, we next assess the magnitude o fthe problem by comparing the average labor costs o f STU operations with those o fprivate operators and estimating the potential cost savings. ReducinP Costs o f STU Stage Carriage Bus Services 2.10. The possiblereduction inunit transport costs due to reducing the staff costs o f State Road Transport Undertakings (STUs) has been estimated by comparing the vehicle operating costs for private and state operators22,and identifyingwhere costs o f state operators could be reduced to the level o fthose pertaining inthe private sector. Only the effects o f staff costs reduction are considered; other potential impacts from STUreform measures are not considered. The global impact is estimated by multiplyingthe possible unit cost reduction, perbus km, bythe recordedbuskmoperatedbythe STUs. 2.11. The vehicle operating costs for stage carriage services inmid2002 have been estimated as shown inTable 2.1 below. These are economic costs reflecting an estimate o f the true cost to India, excluding taxes, subsidies, andother transfer payments. NOTE: (1) Assuming overallcost increaseof4% since 2000-2001for STUs (the average increasefound inKarnataka, Maharashtraand Uttar Pradesh). (2) Assumingfor private operators, the numberof staffper onroadbus is 4.3 (2.5 crew, 0.8 maintenanceand 1.O other) andaverage monthly earningsper staffi s Rs.3,500 (as reportedin PrivatelyProvidedIntercity BusTransportServices, CES, June 2002). Assumingaverage utilization is 100,000 kmper year for stagecarriageoperation. (3) Assumingeconomic conversionfactors describedinTable 10ofthe informal technicalnote on methodology, analysis andreportpreparation. SOURCE: Consultants' estimate ____ 22IanJenkins, et al. "The Efficiency o f India's RoadTransport Industry:Modelling, Analysis, and Report Preparation" (HWTSL InformalTechnical Note for World Bank, May 2003). - 33 - 2.12. Future costs for the STUs have been estimated for three alternative sets of assumptions: 0 Staff numbers are reduced to those levels pertaining inthe private sector but the wage levels remain the same, 0 Staff numbersremainthe same but wages reduce to those pertaining inthe private sector, and 0 Both staff numbersand wages reduce simultaneously. 2.13. The results are shown inTable 2.2. According to this estimate the scope for reducing costs per km i s greatest through reducing earnings rather than reducing staff. The overall potential reductioninpassenger costs i s 38% from Paise 29 to 18 per passenger km. The annual global reduction inbus operatingcosts has beenestimated inTable 2.3 assuming that uniform reductions could be achieved for all the STU operations -both rural and intercity (including hilly regions). Item Current Reduced Reduced Reduced Staff Earnings Staff and Earnings OperatingCost (Rs.per buskm) 9.92 8.16 6.91 6.11 Buskmoperatedper year 10,707 10,707 10,707 10,707 (million) Annual Cost Saving(Rs. d a 18,844 32,228 40,794 NOTES: (1) Assuming staff numbersreduce from 6.6 per bus (4.3 crew, 1.3 maintenanceand 1.O other) to 4.3 (2.5 crew, 0.8 maintenance, 1.0 other). However, current labor lawsmay not permit such alarge reductionin crew numbers. (2) Assuming averagemonthlyearningsreduce45% for all types of staff (from Rs.7,700 to Rs.3,500 on average). (3) Based on recordedkmoperatedby STUs onnon-urbanroutes in 2000-2001 (data coveragewas 94.8%of the fleet so globalbenefitswould be slightly higherthan estimatedabove). (4) According to current plansthe bus km operatedeach year would remain at current levelsinfuture, so potentialsavingswould not change fromyear to year. SOURCE: Consultants' estimate 2.15. From the point o fview o f society as a whole, the way to achieve real net benefits is not byreducing earnings o f staffbut by redeployingthe excess staff into other more profitable forms o f employment. Inthis case, ifthe number o f staff can be reduced as assumed above, the potential annual benefits would amount to almost Rs.19 billion (USD 380 million). However these benefits would not be achieved immediately due to the social problems causedby staff redundancy, pension and/or redundancy compensation costs and the need to respect existing labour agreements, a point to which we return below. - 34 - 2.16. It must benoted that only part o fthese cost savings would berealizedwithin the intercity bus market. Much o fthe bus market served bythe STUs is rural rather than intercity innature. The actual amount is almost impossibleto determine due to the lack o fa clear distinction between the two types o f service. For the three sample states, the proportion o f buses estimated to be operatingintercity services constituted 30% inMaharashtra, 60% in Karnataka and 72% inUttar Pradesh. Ifthis represents a fair cross section o fthe STU fleet, the overall average would be about 50 %. So the potentialbenefits attributable to staff cost reductions inintercity operations would be 50% o f those estimated above. The other 50% would be potential benefits inthe rural bus market. Reform o f the STUs 2.17. The ASRTU has proposed a detailed agenda for reform o fthe STUsector, as summarized inthe Box overleaf. We consider key facets o f that proposal here. 2.18. Generally the thrust to corporatize the STUs as advocated by ASRTU and give them greater management autonomy with greater freedom fkom political considerations inday-to- day management is consistent with classic approaches to the reform o f state owned enterprises. Where there are clearly defined goals and specific, monitorable targets (often in the form o f an annual contract or business plan), and the managers are then actually given greater management autonomy to achieve those targets, this form o f organization can result inimprovedmanagement ofresources andgreater productivity. 2.19. However, experience shows that it i s more often thannot difficult to implement such schemes so long as government i s still the main stockholder, and inrecent years the hrther step o fpartial andthenmajority privatizationinpursuit o f greater efficiency has, o f course, followed inmany cases inmany countries. 23 It i s noteworthy that the ASRTUproposal [Items (b), (c), (d)] is also pointed somewhat inthat direction, envisaging at least divestment o fcentral government equity. Ofeven greater significance is the fact that at least four Indian states no longer have an active STU and instead rely entirelg4on the private sector for provision o f inter-city bus services. That seems the most likely outcome over time for many more states, but inthe interimthere are many steps that can and should be taken during the period o ftransition to improve STUperformance. 2.20. It i s an acceptedprinciple o f good governance, as urgedby the ASRTU [items (n) and (o)] that any government that demands an enterprise meet unprofitable public service obligations should also be requiredto reimburse the enterprise for those services -provided, however, that the enterprise demonstrates that it is an efficient provider o f those services. Accountability i s greatest when the government that demandsthe services i s the one that meets the bill; discipline is lost ifa different level of government (often the center) is expected to meet the obligations demanded by another level (often the local government). That central government no longer financially supports the STUs but leaves this to state and local governments that demand the services i s a move toward better governance. 23Shirley May and JohnNellis, Public Enterprise Reform: TheLessons of Experience (World Bank, Economic Development Institute, 1991); Nasir Islam, "Public Enterprise Reform: Managerial Autonomy, Accountability and Perfonnance Contracts," Public Administration and Development, Vol. 13, pp. 129-152, May 1993; Ian Duncan andAlan Bollard, Corporatization and Privatization: Lessonsfiom New Zealand (Oxford University Press, 1992). 24Private operators are now saidto be using bus terminals and other facilities previously providedfor the STU. - 35 - ASRTU - AGENDAFORREFORM STATETRANSPORT OF UNDERTAKINGS2' Excerptfrom "Policy Issues ConcerningRoad Transport'' The following steps are suggested for improvement in public passenger transport system and a) The RoadTransport Corporations Act may berepealed. b) The State Transport Undertakings under RTC Act, the Municipal andDepartmental Transport Undertalungsmay register themselves as public limited companies underthe Companies Act. c) 1The existing capital contribution (where applicable) made by the Central and State Governments be - - ` converted as equity inthe new Company by allotting shares-to the Central and State Governments. that the region under the company's control i s efficiently and effectively networked and bus Source: www.asrtu.org 25Excerpt from "Policy Issues Concerning Road Transport", By Dr MKoteeswaran, Executive Director, Association of State Road Transport Undertakings - 36 - 2.21. The main difficulty inimplementing such aprinciple - inthis context as inso many others - is indemonstrating that the provider, inthis case the STUs, is an efficient provider. So long as STU costs greatly exceed the costs o falternative providers, as they clearly do, the argument for subsidized capital [items (g) through (i)] reimbursement o f STU losses on for specific services may be viewed as subsidies to continued inefficiency -the tax payers, like the bus riders, would merely be subsidizing above-market wages and excessive workforces o f the STUs. 2.22. At least inthe case o finter-city services, which are inherentlythe most profitable, the case for subsidized routes [item (l)]at best weak; most ifnot all inter-city services should i s be able, and should inpractice, paytheir fullcosts. Our assessmento fthis market is that private capital i s ready, willing and actively investingto meet market demands. There i s not a market failure argument infavor o f government intervention. 2.23. The proposal [items u)through (m)]to classify routes, andbundleroute segments so as to cross-subsidize unprofitable segments from protectedprofitable segments, is unwise, probably unworkable, and inany case unnecessary, inthe case o f inter-city services. It is unwise, first, becauseit is basically a mechanism for generating excess profits to cross- subsidize excess costs byrestricting competition, where the appropriate solution is to open competition to force more efficient solutions; second, it distorts market decisions by exaggerating the costs o f certain services andunderstating the costs o f other services. I t is unworkable inthe absence o fvery tight constraints on market entry -as demonstrated countless times over the years inmanycontexts -as it creates strong incentives for cream- skimmingcompetition for the mostprofitable services. That typically leaves the licensed franchisee saddled with a heavy burden o funprofitable services andthe government saddled with aheavy regulatory burden. Not least, inthe case o finter-city services it is simply unnecessaly because the evidence i s that those services are sufficiently profitable to support unrestrictedcompetition. Nor i s there an economic rationale to justify forcing inter-city travelers to help pay for either rural feeder services or urbanservices. 2.24. The proposalthat the STUs should act not only as an operator but also as the government regulator [items (k) and (q)] would create an egregious conflict o f interest and an unacceptable moral hazardwith the prospect that the STUwould act first to protect itself from unwanted competition on the profitable routes. The present circumstances are that there is too little competition for the stage carriage routes, many o fwhich remain nationalized, i.e. the legally protected monopoly o f STUs. 26 26 In this context, it i s interestingto note that inKarnataka and Uttar Pradesh the proportiono f total market offerings o f seats offered by the STUs vs private operators has beenhigher on the profitable inter-city services (respectively 22:78 and 26:74 percent) than on the rural services (respectively 19231and 14:86 percent). O fthe three statesreceiving particular attention inthis study, only Maharashtra STU offers a higher percentage o fthe market inthe rural segment (54:46) than inthe inter-city segment (39:61). Consulting Engineering Services (India), "Privately Provided Inter-city PassengerTransport Services," Report to the World Bank (June 2002). - 3 7 - Competition, Restructuring, and Privatization 2.25. Iftheprimaryobjective isthebest quality service at thelowestpossiblecost, the long-run strategy will be to move the STUs to majority private ownership incompetitive markets as early as possible. Arguably, competition is more important thanprivatization. However, so long as there remain large labor forces employedbypublicly owned camers with highunit costs, the political andmarket pressuresto restrictcompetitionto keep tariffs highto protect the least efficient carrier willremainstrong, andthis willprotectnot only the STUs but also serve as an umbrella to protect marginally efficient firms inthe private sector as well -all at the expense o fhigher fares for bususers or higher taxes for the citizens at large. 27 Incontrast to the idealistic motives o fthose who helped create the STUs -and undoubtedly manywho are still engaged with them- the public investment that was intendedto provide improvedtransport for a generally disadvantaged public has insteadbeen diverted primarily to the benefit o f the relatively advantaged workforces o f the STUs. 2.26. Given the magnitude o fthe labor engaged by the STUs2*-their total labor force numbers more than 700,000 -and their present unprofitability, privatization will not be accomplished overnight. However, delineation o f a clear restructuringplanto transition toward that goal with the fillparticipation o fthe major stakeholders can provide essential foundations, allay unwarrantedfears inthe short run, and facilitate orderly movement toward the ultimate goal. 2.27. A central element o fthis planwill be a labor downsizing, retraining, andre- employment program. This can be accomplished over time bynatural attrition, by not replacing employees as they retire or leave for other employment opportunities; by offering early retirement incentives that accelerate the process; byjob retraining (either for outplacement or to fill differentjobs within the STU as vacancies develop); and by offering improved termination grants. Such measures can assist both incushioning the personal impact on the affected employees andinproviding well equipped labor to meet the broader economy's labor needs. 2.28. Bynow there has beenvast experience with this sort o fenterprise restructuringsand privatizations around the world, 29 and the World Bank has been involved ina large number o fthem. It can provide not only technical assistance for planningthese activities, but also in certain circumstances capital finding for retraining and re-employment aid programs, early retirement schemes, and other measures to assist inthe transition. 27Christopher D.Foster, Privatization, Public Ownership, and the Regulation of Natural Monopoly (Oxford UniversityPress, 1992). '*Some o fthe larger STUs have more than 100,000 employees; for example, Maharashtra State RoadTransport Corporation (MSRTC) reports 112,116 employees. 29See, for example, DavidH.Fretwell, Jacob Benus,and Chstopher J. O'Leary, "Evaluating the Impact o f Active Labor Programs: Results o f Cross Country Studies," World Bank Social Protection Discussion Paper No. 9915 (June 1999); Gopal Josh (ed), Privatization in SouthAsia: Minimizing Negative Social Efects Through Restructuring (International Labour Organization, Geneva, 2000); Kaushik Basu, Gary S. Fields and Shub Debgupta, "Retrenchment, Labor Laws and Government Policy: AnAnalysis with Special Reference to India," Cornel1University (undated); andAntonio Estache, Jose Antonio S c h t t de Azevedo, and Evelyn Sydenstricker, "Labor Redundancy, Retraining and Outplacement DuringPrivatization," World Bank Institute PolicyResearchWorking Paper 2460 (October 2000). - 38 - 11.2 PrivateInter-City BusTransport 2.29. Inrecentyearstheprivatesectorhaswonbackarapidlyincreasingshareoftheroad passenger market, and ina few states private operators have already entirely supplantedstate- owned services; only Andhra Pradesh reportedly has as yet no substantial private bus competition to the STU. Private bus operators have also established a new standard o f service inlong-distance inter-city services (particularly inSouthern India), while the market for shorter-distance transport i s beingtransformedbythe introduction ofmodemsmall to medium(10-18 passenger) buses, which operate more efficiently onthe ruralroutes. Inother states, a policy o fhiring private buses by STUs to supplement their fleet strength and to operate under STUmanagement on still-nationalized routes has recently beenintroduced. Altogether, from a low o f about 55 percent o fthe bus fleet in 1981-1982, by now more than 80 percent o f the nation's bus fleet is reportedly inthe hands o fprivate owners. 2.30. However, far less i s known about the private bus carriers who are now thought to meet about three-fourths o f the bus transport demands inIndia than is known about the STUs that provide about one-fourth. That i s partlybecause, incompetitive markets without market distortions, the economic characteristics o f the industry-few economic or technical barriers to entry and few economies o f scale to firm size -dictate an atomistic industry comprising manyvery small supplierswho have little need for complex management information systems -incontrast to the huge state owned enterprises that comprise the STUs, which came about solely because of government intervention. Confidentiality o f information insuch competitive marketsis also normal management practice. Studyo fPrivate Bus Transport inKarnataka, Maharashtra, and Uttar Pradesh 2.3 1. To address the gap inavailable information, the World Bank in2002 commissioned Consulting Engineering Services (India) Pvt Ltdto carry out a study o f private bus operations inthe statesofKarnataka, Maharashtra, andUttar Pradesh. The study3'includedprimary surveys encompassing 63 bus operators and4,302 bus users; inaddition 128 passenger terminals inthe three states were surveyed. The study was confined to inter-city andrural bus transport; urban transport was excluded. A summary ofthe findings is given below: 0 Inbothrural andinter-city segments ofthepassenger transportmarket,stagecarriage operations are still predominant. Restrictions are reportedwith respect to grant of permits, both innationalized and non-nationalized areas, yet the STUs neither expand their fleet nor generally officiallyallow private operators inthe nationalizedareas. Although recently, insome cases private buses are being recruited under STU management under the km scheme. 0 To bridge the gap between demandand supply, clandestine operations has thus become a rule rather than an exception. Inreality the markets are increasingly competitive. Most o fthe private buses havingContract Carriage permits violate the permit conditions and operate as stage carriages. They also have to compete with the multitude o f smaller vehicles such as maxicabs, jeeps, vans and LCVs andtourist cabs, all o fwhich operate 30Consulting Engineering Services (India), "Privately Provided Inter-city Passenger Transport Services," Report to the World Bank (June 2002); hereafter cited "CES Study". - 39 - without any regardto the permits they hold. These small vehicles do not follow any regulations either infares, on routing matters or intimings o foperation, and their operations are fully flexible and vary instantly with traffic demand. Insum, where the STUs have not met market demands, the market i s working around the failures o f the government instruments-the STUs and the restrictive policies designedto protect them- to meetsocietal needs (albeit illicitly). Incertain states, both maximum and minimumfares and in other statesspecificfares are fixed, having due regard tophysical terrain, road surface, type of service, area of operation... and volume of trafic. It is, however, unclear to what extent fare regulations are observed; the CES survey reported conflicting results. Probably the true situation i s a mixedpattern, with the degree of adherence to statutory fares depending on how closely those fares happen to matchthe market equilibriumprices at the specific time andplace. The most striking findingo fthe CES study is that by a rather large majority bus users in the three states now appearto be reasonably satisfied with current bus services, as offered bybothprivate operators andthe STUs, althoughnoticeablyless satisfaction was expressed inUttar Pradesh incertainrespects. Between 63 and 70 percent o fthe users considered the vehicle condition and overall quality of service satisfactory or good; only intwo districts ofUttarPradeshdidas many as one-third consider itbad. As tofrequency of service, more than 90 percent inKarnataka and Maharashtra and 70 percent inUttar Pradesh considered the situation satisfactory or good. On seat availability, more than 90 percent inKarnataka and Maharashtra considered the situation satisfactory or good, and inUttarPradeshthe corresponding numberwas 80percent, although 16percent considered it bad. Fares were rated as satisfactory or good by over 90 percent o fthe users inallthree states. With respect to terminalfacilities, inMaharashtra 82 percent and inKarnataka 61 percent o fbus ridersratedthem good or satisfactory, but inUttar Pradesh, hlly 69 percent rated them bad and only 10percent considered them good. Typically private operators are not givenaccessto the mainterminals, which are controlled bythe STUs. With respect to users'opinions about competition, they generally viewed competition favorably as it has resultedinmore frequent service, more choice o f modes (stage, contract, and tourist bus services, maxicabs and taxis) and a wider range o f level o f comfort to choose, better availability o f seats, and more competitive fares. 2.32. One mustbe cautious in extrapolating from this brief survey o fpassengers inthree states to the rest o f India. However, we do know that markets normally work well when there i s substantial competition, and less well where competition i s limitedfor whatever reason -whether due to government restrictions on entry or fares, or natural economic factors that limit competition, such as thinmarkets where revenue potentials are less than costs o f service provision, even for the most efficient producers. Fortunately, there is a wide array o f technologies available to adapt to the characteristics o fthe diverse markets for road passengertransport -which vary spatially, temporally (by the seasons o f the year, the hours ofthe day, andby degree of comfort (availability o f seating, air conditioning) for those who can afford it. As noted above all o f these (large, medium and small buses; maxicabs, cars, - 40 - trucks; standard, deluxe, super deluxe, air conditioned classes) are beingdeployed in the market to meet the demands o f specific micro-markets. 2.33. We would therefore expect to findmuchthe same conditions throughout Indiawhere the scale o fthe market is generally sufficient to attract competitive private providers o f one form or another, except where significant, persistingmarket distortions -either those introducedand enforced by the Governments or by the efforts o fprivate parties to twist the markets to their own advantage. 11.3 RevampingGovernmentPolicy to Better Serve the Public Interest 2.34. Governments have multiple (oftentimes conflicting) objectives and are universally constrained byresource availabilities. Inthe case ofpassenger transport, equity considerations may weigh as heavily, or possibly even more heavily, than economic efficiency. Equityconsiderations inthis case include the concept that bus riders throughout a state (or particular sub-region thereof) should pay the same fare regardless o f costs o f providing the specific service3' and also the concept that certain groups insociety (e.g. retired persons, school children, militarypersonnel) should enjoy significant fare discounts. It is reasonable to infer that such equity considerations were key factors inthe original decision bythe GO1to establish the STUs in 1950, andthe concomitant decisionto restrict licensing for stage services as a monopoly for the STUs inorder to create profitable routes that could cross-subsidize the unprofitable routes where the standardfare was infact below the costs o f service. 32 2.35. Such structures, which were quite common around the world at one time, have generally proved difficult to sustain, as the market disequilibrium thus created offers a very great temptation to private providers to enter the market incream-skimming competition. As notedabove, that has typically left the licensedfranchisee saddled with a heavyburdeno f unprofitable mandatory public services andthe government saddled with an ostensibly heavy regulatory enforcement burden-that inIndiahas, o f course, as we have seen, been circumventedby the market, undoubtedly aided from time to time by illicit payments to encourage enforcement officers to look the other way. 2.36. From an economic perspective it i s fortunate that the market has managed to circumvent these restrictiveregulatory policies, as the result is a more flexible transport system, moreresponsive to the specific needs o fthe passengers andthe Indian economy. However, from the perspective ofpublic governance, having government regulations flaunted so openly undermines the public's respect for law and order more generally. 2.37. Inthis context, the appropriate public policyprescriptionis to remove quantitative economic regulations restricting market entry, and instead allow markets to determine what 31As inthe bus official fares fixed byKarnataka and Maharashtra states, which are ostensibly applicable across the entire respective states. The fares fixed by Uttar Pradesh state at least attempt to reflect terrain and road condition. 32The policy o f IndianRailways inkeeping railway fares far below eventhe variable costs o f provision-to the extent that the railways are clogged with unprofitable passengertraffic and have had to forfeit other profitable traffic which the railway would be expected to cany far more cheaply than the modes now canying it- also suggeststhat equity considerations o f a kindinfact weigh more heavily to the Government o f India in determining passengertransport fares than efficiency considerations. -41 - services will be offered. At the same time, qualitative regulationsto ensure safe, reliable services, and environmental protection, should be further developed and seriously enforced. Inthis manner the government couldbest ensurewell functioning markets that provide the array o f services the various market segments demand and at least cost. Deregulation and Facilitation o fPrivate Sector Entryand Competition. 2.38. With a view to creating a market inwhich passenger services o fvarious types and size compete with each other, unassisted, policymakers inIndia should be concerned with puttinginplace aproper regulatory environment, and expandingthe role o fthe private sector within this environment. Inparticular: (i) regulations that internalize social costs, such as those related to the environment, safety and congestion, so that the market can allocate resources ina socially desirable way; and (ii) regulations that establish basic rules for fair competition should be developed and implemented. The requiredreforms and changes in regulatory policy, necessary to promote public safety as well as low fares, have not yet taken place in all states. Without reform, private sector finance, maintenance and operation o f passenger services i s likely to continue with some undesirable consequences. 2.39. Introducing Fair C~rnpetition~~. There are many ways o f introducing fair competition inservice provision to the inter-city passenger transport markets inIndia. Route franchising is a means o fmaintaining some public control over the level o f services and prices inthe public passengertransport market, while usingcompetitive forces to secure supply at the lowest cost. This can apply to non-remunerative bus services alone (as inmost o fthe UnitedKingdom) or for all services (as inLondon and inCosta Rica); with the supplier either carrying only the cost risk(as insome cases inthe UnitedKingdom) or carrying boththe cost and revenue risk (as inNew Zealand). Where fragmented competition i s not possible because o fthe indivisible scale o f operation, market disciplines can still be employed by competitively concessioning facilities or systems. This has been applied to the management o f urbanbus systems, particularly infrancophone Africa. 2.40. Competition between groups within a licensedfranchise system can be promotedby ensuringthat the routes for which monopoly franchises are granted overlap sufficiently to encourage competition for customers on common sections o f route. This approach is practiced to secure competition between different bus operators' associations inBuenos Aires and other LatinAmerican cities andbetweenoperators o fdifferent kinds o fpublic transport vehicles insome African countries. This form o f competition makes it possible to some degree to organize supply, and limits anti-competitive operatingpractices, as long as there i s a competent franchising authority to prevent the emergence o f a single strong cartel. 2.41. Competition between modes can be effective where demand is dense and varied, as exemplified by the role o fprivately operated minibuses inHongKong and Dakar. Some flexibility towards the introduction o f new categories o f services at higher prices maybe a means o freconciling the maintenance o f a basic low fare with the provision of adequate total capacity and a sufficiently varied range o fprice/quality combinations to meet demand. Withinregulated systems, this can arise by design as inthe "two-tier" bus systems inSeoul or Shanghai. A system allowing a range o f services to be provided legally bythe informal sector may be less susceptible to the sprouting o f illegal services. - 42 - 2.42. The Casefor some Regulation. There are two reasons why it may be desirable to retain some public regulation of the right to supply public transport inthe inter-city bus markets. First, regulationmay be desirable insome cases where an unregulated market process may result in: (i) matching o f schedules (inlocal busmarkets); (ii) increased pressure to engage indangerous practices, such as racing o fbuses (often increasingaccident rates as with the "red buses" inDelhi); and (iii) perceived losses inthe stability and reliability o f service, with losses incustomers andreductions invehicle occupancy, as occurredinthe bus markets insome British areas after deregulation. Second, while cost reductionsresulting from unfettered competitionmay allow previously unprofitable services to continue, and may even leadto more frequent services beingprovided on previously non-remunerative routes by using smaller vehicles that are more suited to low demand, sometimes social objectives may require direct financing o f some services that might otherwise be lost through competition in the market.For example, the elimination o f cross subsidies may reduce supply or increase the prices o f services affecting the very poor, as inthe case o frural bus services inSri Lanka. In such situations, making markets "contestable", through competition for the right to provide subsidized services at least cost, will still allow non-remunerative services to be provided at the least real cost. All o f these defects o fthe market process may require qualitative controls butnot necessarily monopoly franchises andnever direct state involvement inservice provision.34 Monitoring for Anti-competitive, Cartel/monopoly-type Behavior o f Companies 2.43. Inanunregulated market,profitmaybesought throughthe creationofanoperators cartel, as occurred inthe bus industryinSantiago, or by operators combining with suppliers of terminals or other infrastructure to exclude competitors from access to crucial facilities. Controlling anti-competitive behavior requires a regulatory institution to prevent the acquisition and exploitation o f excessive marketpower.35Inpractice, the regulation o f cartels is not a simple task as some forms of combination, such as operators associations inpublic transport (for example, inBuenos Aires buses) may actually contribute to the efficient workings o fthe market.36Evenwithout cartels, wherever there i s a financially strong incumbent ina market, there i s a danger that anti-competitive predatory behavior will occur. The most efficient markets for road transport operations normally comprise very large numbers o f very small producers, and government policies should inany case avoid trying to bias market determination o f firm size. Only if,contrary to expectations, some firms or associations o f firms, grow so large as to threaten the competitiveness o f specific sub- markets, i s it necessary for anti-monopoly authorities to intervene. 2.44. Also inthe case o f India, since larger firms cannot be expected to be more efficient providers inroadpassenger transport, it maybe advisable to accompany deregulation with a restructuring of STUs into a number o f smaller firms to curtail their market power. Ownershipcouldbe fragmented ifnecessaryby sharing ownership o f assetswith employees. 34Darbera, R. 1993."Deregulation ofUrbanTransportinChile: What havewe Learnedinthe Decade 1979-1989?"Transport Reviews 13 (1): 45-49. 35Henry, E.andR.S.Pacheco. 1994."Relations depouvoirsentre entreprise dautobuset tutelle:re fltxions A partirducas du Brtsil" inX. Godard(ed). Les transportsdans les villes du Sud. Paris: Karthala. Quoted inWorld Bank, Sustainable Transport, 1995. 36Armstrong-Wright,A. and S. Thiriez. 1987."Bus Services: ReducingCosts, RaisingStandards"WB TechnicalPaper 68, World Bank,WashingtonDC. QuotedinWorld Bank, SustainableTransport, 1995. - 43 - Competition would then be subject only to general oversight by the competent national authority responsible for ensuringfair competition. 11.4 RecommendedPublic Policy for Inter-City Bus Services 2.45. The specific steps recommended are detailed inTable 2.3, and summarized below : (a) To establish atrue level playing field for competition, it will benecessary to restructure the State Transport Undertakings as outlined above to make them far more efficient, andprobably to divide them into smaller units to curtail their market power. (b) Monopolyrestrictions on stage routes, including current restrictions on interstate stage carriage underbilateral agreements, should be removedby allowing free entry; only technical qualifications relatedto safe provision o f services should be required. (c) Terminal facilities shouldbe opened to private operators on a pay-for-use basis. Where the capacity o f these facilities is inadequate, a mechanismfor periodic auctioning o f slots could be institutedby an agency independent o f the STU. (d) Motor vehicle tax should be imposed on a common basis. The present system imposes passenger tax (as a portion o f fares) on formally certified operators and increases their marginaltax rate above that o fthe public sector (lumpsumper vehicle tax). Apractical approach would be to tax all operators on a per vehicle basis. (e) Some vigilance will need to be exercised against the efforts o f either the restructured STUs or largerprivate carriers to monopolize the market or support formation o f carrier associations or cooperatives that become cartels inspecific market segments. The independent agency referred to in(3) above, could also serve this function o f ensuring competition, preferably inthe market, andwhere that i s not feasible, by concessioning franchises that are contested periodically. 2.46. Inthe absenceof any strong economies of scale to firm size, the dangers from monopolization or cartelization inthe sector arise primarilyfrom the efforts o f the STUs to maintaintheir privileged, protected position or inappropriate government policies that would seek to lendmarket power to larger firms, cooperatives, or cartels than they would naturally have. Inthe CES Study in2002, just as inthe 1999 landmark Sundar Committee report on Trucking Operations in India -and also as inIndia's various Five Year Plans from the beginningmore than 50 years ago -there is a markedpreference by Indiananalysts and policymakers alike for the government to stimulate the formation o f larger companies inthe bus andtrucking industries. That is apparently due inpart to the perceptionthat there are substantial economies of scale to firm size inthese industries -a point that is not generally sustained inthe wider economics 1iteratu1-e.~~(The preference would appear at times also to 37The risk o f market distortions due to the presence of scale economies relates to the structure o f the bus industry. Some analysts note that after UK's bus deregulation, there was a significant reconcentration o fthe bus industry after an initial period of expanded entry, accompanied by fare increases. Ths was attributed to the hub-and-spoke structure o fthe UK bus industry developed under regulation which gave the large network carrier cost advantages over independents, plus the ability to foreclose competitors by providing inferior terminal access. By contrast, the US bus industry, although characterized by a single nationwide private carrier and numerous small local and regional carriers, has seen l o w prices and frequent price wars. This difference in - 44 - be due inpart to the perception that larger firms are better organized and would provide government policymakers with better, more extensive information on which to base their policies -which, iftrue, would constitute an inversion o fthe proper relationship o f policymakers to the market.) Table 2.3: Recommended Policy actions for Inter-city Bus Services Area Policv/Reform Actions Promote corporatization or Develop a restructuring and commercializationprogram for commercialization o f STUs STUs and diversification o f D Develop a labor downsizing, re-training andre-employment ownership program for STUs Explore mechanisms to provide financial support for fleet modernization2 8 Eliminate subsidies (direct/ indirect), to STUs Create an independentboard o f directors for each STU to hold management accountable; i DiversifySTUownership' : -Offer stock ownership to employees - -Convert long-term bank loans into stock Formulate and implement 8 Draft guidelines to indicate trade practices regarded as competition policy unfair or anticompetitive Draw up regulations to restrict unfair practices and anticompetitive activities 0 Select agency responsible for monitoring and enforcing competition rules and ensuring access to common use infrastructure facilities. 0 Compile price information based on the tariffs filed (by operators) and disseminate aggregate data to the public so the market can function more efficiently Establisha more flexible 0 Remove monopolies on state routes regulatory framework so bus 0 Implement a uniform tax system on a per vehicle basis operators can respond to 0 Modify existing regulations to deregulate fares3 changingmarket conditions 0 Review current vehicle inspection system and coordinate more swiftly. efforts so as to eliminate any duplicated regulations/facilities 0 Implement regulation concerning safety and environmental impacts Notes: (1) In some areas where immediate corporatization or commercialization may not be feasible due to the inadequate safety net and lack of legal framework, diversification o f ownership should be pursued. (2) This would include encouraging the creation o f financial instruments (such as leases and equipment trusts and certificates) designed to provide the industry with private capital. outcomes may reflect the linear organization o f the U S bus industry, where most routes are single buspoint-to- point services. With few scale economies, regional competition from small carriers along individual routes i s intense and there are few indications that market power by the largest carrier i s o f any concern. See Richard G.Sharp, et al., "A Cross Country Comparison o fRegulatory Reforms to Promote Road Transport Efficiency" (HWTSL Informal Technical Note, May 2003). - 45 - (3) Shifting the setting o f rates from the Government to a system where operators set prices and file tariffs with transport authorities. The latter would take action only when tariffs grossly deviate from the rates on file and use their market power to discriminate against specific customers. 2.47. The appropriate focus of regulatorypolicies inthe case o f inter-city bus services should be qualitative standards related to ensuring, first, the safety o fthe services, and, second, the minimization o fnegative environmentalimpacts. Safety dimensions encompass vehicle road worthiness standards (brakes, steering, tires, visibility, lighting and signaling), driver qualifications and working hours, and avoidance o f excessive overloading (ridingon the outside or top of buses as happens from time to time inIndia i s not conducive to safety). Unfortunately, inIndia these beneficial regulatory dimensions are also not generally enforced for the same reasons that economic regulations are not generally enforced, i.e. transport operators generally find it more advantageous to make "facilitation payments" to the transport authorities. Incontrast to quantitative regulations, qualitative regulations can contribute greatly to improved safety and environment. They shouldbe seriously enforced. - 4 6 - 111.COMMERCIALMOTOR INSURANCE 3.1. This section examines the role playedbythe commercial motor insurance industry in facilitating safer drivingpractices andimproving the overall safety record on the Indian roads. The main premise underlyingthis i s that the effectiveness o f the commercial insurance market inimproving road safety rests on the ability to differentiate between "good" and "bad" risks by providing strong economic incentives for betterhafer drivingthrough differential premium rates, varying deductibles, and outright refbsal o f coverage to reckless and dangerous drivers. Insurers can also serve a valuable role inresearching motor accidents and supplyingrelevant informationto the authorities. 3.2. The insurance companies' ability to differentiate between "good" and "bad" drivers however is contingent uponthe availability o f adequate easily verifiable information on the insured drivers' past drivingrecord, work experience, training, type o fvehicle driven, area o f operation and other socio-demographic characteristics. Inaddition, the ability o f an insurance company to provide fairly priced commercial motor insurance cover depends on fast and accurate reporting o f claims and the quality o f legal process guidingclaim settlements. Any breakdown inthe above described information flow, or inlegal or institutional arrangements will create price uncertainty, thus resultinginrationingo finsurance coverage either through higher prices or lower quantity o f coverage offered by insurers. 111.1Current Stateof the Indian CommercialMotor Insurance Market38 3.3. India is among the most road accident prone countries inthe world. While accounting for only 2 percent o fthe global car fleet, it i s responsible for over 7 percent o f all accidents world wide, which result inover 85,000 deaths and 250,000 injuries per year3'. Over 70 percent o f these accidents are contributed by commercial vehicles, buses and trucks, despite the fact that they account for only 5 percent ofthe total car fleet. The rate o f fatal accidents on the Indianroad i s one o fthe highestinthe world, 20 deaths per 10,000 vehicles, which is over ten times of that inHolland and the UK.The overall social and economic costs ofcar accidents are about 2 percent o f GDP40. Business Fundamentals o f Commercial Motor Insurance 3.4. Motor insurance ingeneral accounts for about 40.68 percent o f the gross insurance premiumwritteninthe Indiannon-life market, or over USD 1.4billion, which makes it the largest line of business today. Most o fthat business i s written by the 4 public sector companies. While no accurate statistics are available on how much o fthat premium should be attributedto the coverage o fcommercial vehicles, according to some expert estimates, premiumfiom commercial auto policies is onthe order o f 50percent which explains the importance o fthis business line for the Indianinsurance market as a whole. 38Fromhere onwards the term "commercial auto insurance" pertains to insurance coverage for trucks (both heavy andmidsize vehicles) andbuses drivenprofessionally for commercial gain. 39Source -Loss Prevention Society o f India. 40TheWorld Report onRoadTraffic InjuryPrevention byWHO andthe World Bank, 2003. - 47 - 3.5. For years, the economics o f the motor insurance coverage inIndiahas been known for its poor and ever deteriorating fundamentals. This conclusionhowever should be qualified by distinguishing between the "own damage" (OD) andthe "third party liability" (TPL) part o f coverage. While the former is mildly profitable, with the loss ratios averaging around 60-70 percent, the latter is a lague on the Indian insurance industry, with loss ratios for the industryapproaching 125%4? The loss ratio for the commercial TPL, however, i s considerably worse, ranging between 220% and 500%, which by far makes commercial TPL the most unattractive line o fbusiness inthe Indianinsurance market. 3.6. To provide commercial auto insurance cover, insurance companies have to use a standardinsurance policy form authorized bythe IndianInsurance Regulatory and Development Authority (IRDA), which contains the following two main coverages: (i) own damage to the vehicle; and (ii) thirdparty liability (TPL) to cover claims arising from damages caused to property, health and life o f third parties. The cover contains no deductible and is introduced on the first loss basis. The TPL cover i s compulsory for all drivers and can be bought separately. Unfortunately, the level o f compliance with this requirement i s quite low. While no detailed statistics are available, the surveyedinsurers indicate that the level o f insurance penetrationfor the TPL coverage is well under 50 percent o fthe driver population. The TPL cover i s a direct extension o f the Motor Vehicle Act o f 1988.42 Insurance TarifP3 3.7. One o fthe mainreasons cited by the surveyedinsurance companies for the growing losses inthe auto segment o f their portfolios is the motor insurance tariffmaintainedby the IRDA for bothOD andthe TPL partso fcoverage. Tariffs, per se, reduce the scope o f competition, bringrigidity inthe market, discourage adaptability to the changing needs o fthe insurance consumers and deprive the market o f i n n ~ v a t i o nDue . ~ ~to the fierce competition andpolitical considerations, the "minimum" tariffhas effectively become the "maximum" tariff charged by the state-owned companies. 3.8. Besides mounting financial losses o f state-owned companies, the tariff regime discourages state-owned companies from engaging inproper screening and underwriting o f motor risk. Indeed, at the moment any improvements inrisk underwritingor investments in risk reductionby state insurers are unlikely to yield them any major benefits as long as they 41Loss ratio i s a relationship o f incurred losses to earned premiums. For instance, loss ratios inexcess o f 100% are considered loss-making as they meanthat an insurer pays out more inclaims that it collects inpremiums. 42 The mainpurpose o fthe TPL compulsory requirement i s to provide a financial safety net to the dependents and victims o f road accidents. The TPL cover addresses the natural consequences o f a road accident to the victim such as trauma, injury, and if he is employed, loss o f earnings. Incases o f fatal accidents, the TPL cover compensates the dependents o f the deceased for medical and funeral expenses, loss o f consortium and loss o f estate as well as income streamthat would have benefited the family. 43 The TariffAdvisory Committee (TAC) was establishedby an amendment to the Insurance Act, 1938 (effective June, 1969) to control and regulate the rates, advantages, terms and conditions that may be offered by insurers inrespect o f any risk or any class or category o f risks inthe general insurance business. The non-tariff products are filed by the insurers with the Authority under file anduse procedures. In2002-03, non tariff business constituted about 25 per cent o f the gross direct premium underwritteninthe non-life segment. 44IRDAAnnual Report, 2004. - 48 - haveno choice but to provide coverage at a fixed price to each and every customer applying for cover as a part of their social mandate. As a result, under the present regulatory regime, the state-owned insurers are unlikely to become the major driving force o f change inthe area of risk management and road safety. 3.9. Inadditionto the low tariff, the claims settlement process also imposes significant burdenson the industry.Dueto the highfrequency of car accidents inIndiaandthe particularities of the country's legal system (see Section II), the courts are overloaded with claims. As a result, on average it may take 3-5 years for a case to settle from the day it i s filed. There are cases however which take much longer. 3.10. Due to the long-term claims settlement process, the TPL i s viewed as a "long-tail" business for reserving purposes. This means that an estimated amount o f indemnity plus accrued interest penalty (awarded inthe case o fIndia) has to be put aside inloss reserves once the accident has beenreported. An additional amount has to be added to the incurred butnot reportedloss reserves as well to provide for potentialupwardrevisions ofthe claim estimate inthe future. These reserve amounts have to be revised annually based on the progressiono f the case. The state-owned insurance companies however simply do not have the internal technical or humanresource capabilities to fallow proper reserving practices. The main andmost likelyimplication o f this is considerable under-reservingwhich potentially can have a major impact on the companies' solvency. 111.2 StructuralImpedimentsto an EfficientIndianAuto InsuranceMarket 3.11. The state o f the existing legal framework, deficiencies inthe process o f generating accident data and data sharing, and the enforcement o f safety regulations inthe motor transport industry are the most serious structural impediments to the operation o f the Indian commercial motor insurance market. Each o f these issues is discussed below. Legal Impediments 3.12. The legal framework guidingthe operation o f the Indianinsurance market by and large i s based on the Motor Vehicle Act (1988), and numerous Supreme Court Rulings on specific insurance related cases which provide for: a Unlimitedfinancial liability of insurance companies. The main negative side effect o f the unlimited liabilityclause is that it encourages litigation andresultsinan unusually longclaims settlement process, which may take up to 8-10 years to settle. This puts a major strain on the poorer victims or their families as they have to front the costs o f the accident to the family inanticipation o fthe court award. a No Statute of Limitations. The MVA envisages no time restrictions onperiod within which a TPL claim canbe filed since the occurrence o f an accident. As a result, it takes on average 18 months for a claim to be filed, which opens wide possibilities for fraud and results inmajor financial uncertainty for the insurers. a Veryfew limited legal defenses for insurers. Section 149 o f the MVA provides for only three legal defenses that can be assumed by insurance companies to deny claims. - 49 - Inrealityhoweverthesedefensesweresubsequently severely limitedbythe Supreme Court rulings, which in general took the position that the paramount role o f insurance is to lessen the burdeno f social hardship regardless o f accident circumstances thus effectively endorsing the concept o f absolute liability by the driver. Despite the fact that special commercial Tribunals were set up inevery state to try legal suits arising from motor accidents, the sheer volume o f suits quickly overwhelmed the institutional capacity o f the Tribunals. One o fthe reasons for the growing claims backlog seems to be the financing o f the Tribunal services. Currently, the Tribunals are financed by the Central Government since they are viewed as an extension o f the overall Indian legal system. No filing fees or any other charges are imposed on the suing parties onthe grounds o f maintainingthe Tribunals' independence. As a result, the government budgetaryresources allocated to the Tribunals are very scarce and certainly do not provide for a major expansion of the system, recruitment o fnew staff, and computerization o fthe courts' records. Information Flow 3.13. Insufficient information and inadequate means to verify it are at the heart o f the problems plaguing the Indiancommercial auto insurance market. Those include the integrity of licensing andregistration systems operating at the state level, quality o f drivers' training, training o f local and traffic police inproper recording, filing and forwarding o f accident reports, and the absence o fintegrated countrywide databases on commercial drivers that could be accessedbypolice, courts, insurance companies, and truck owners. Safety RePulations and Enforcement o fthe Motor Transport Industry 3.14. Although there are rather strict requirements for commercial drivers their enforcement i s highly suspect inthe absence o f national integrated databases on drivers' accident record and the ease with which licensing and registration procedures can be circumvented. This laxitywith drivers' professional certification andthe lack o ftraining facilities are among the keyreasonsbehind a very highrate o f accidents for commercial vehicles. 3.15. Inahighlycompetitiveindustrysuchascommercialtrucking, andintheenvironment o fno or minimumlegal or social protection, the cost o f labor remains under constant pressure to be revisedeven further downward, frequently to the detriment o froad and driver's safety. Many drivers have multiple driver's licenses issuedto different names, which inthe absenceo fthe nationalidentification card, are very hardto verify. Today, the driver's safety record andhis marketability are not correlated. Inthe case o f vehicles, vehicle permit data are frequently unreliable and the driver's details recordedinthe permit are not necessarily o fthe person at the wheel at the time o f an accident. 111.3 PolicyRecommendations 3.16. Establishing an effective motor insurance regime is going to be a complex undertaking, but definitely not one beyond the realms o fwhat is possible giventhe current situation and the institutional infrastructure on which India can build.Our recommendations outlinedhere can clearly not be implemented at one go, but will need to be adapted and phased inaccordance with the particularities o f the Indiansituation. These recommendations - 50 - are therefore intended much more as objectives for reform rather than a blueprint to be acted upon. The next step ought to be to devise a plan that i s feasible and acceptable to the various stakeholders inthe system. Commercial Driver's Licenses and Vehicle Inspections 3.17. Getting a Driver's License. It is o fparamount importance to tightenup licensing requirements for issuingcommercial drivers' licenses, put inplace a "points" systemwhich records the accident history of the driver, and establish a country-wide network o f specialized training facilities for commercial drivers. 3.18. VehicleRegistration and Inspection. Giventhe poor enforcement o f vehicle registration and inspectionrequirements, Indiamay consider international experience in registration o f commercial vehicles with the view to tighten up the vehicle registration process and improve road safety. Inthat context, the GO1may consider contacting the International Road Transporters Union (IRU)about modem vehicle registration systems for commercial vehicles. Motor Insurance for Commercial Vehicles 3.19. Detarf$ng the OD and TPL.The existing tariffregime for the OD andTPL appears to be among the key obstacles to the development o f a full-fledged commercial motor insurance market. Detariffing boththe OD and TPL parts o fcoverage would become an important step toward improving the financial viability o f the auto insurance market and restoringthe essential role o f insurers as risk managers andcommercial enforcers o f safety on the roads 3.20. TransparentFinancial Reporting. It is recommendedthat the existing financial reporting standards for insurers be modified by the regulator to provide for a more detailed breakdown o f companies' premium income, with separation o fmotor business premiuminto a separatereportedpremium category. 3.21. Claims Data and Resewing Practices. Recording andreporting o f claims remains one o fthe main challenges faced by all four state owned insurers. Inthe absence o f adequate IT systems and qualified computer literate staff, motor claims are reportedby branch offices with major delays andclerical errors are common. Itis thusrecommendedthat the IRDA jointly with the Tariff Committee commission an independent claims audit inone o f the state owned firms to establish the true picture with regardto (i) the companies reserving practices; and(ii) adequacy of the current tariff giventhe companies' real loss ratios. Such a claims audit can also shed light on internal claims reportingpractices instate-owned insurance companies. 3.22. Integrated Claims Databasefor All Insurers. There is a strong need for creating an integrated insurance claims database that can be sharedjointly by the Indian insurance market. Such a databasewould enable insurers to limit adverse selection and the possibility o f "bad" drivers taking advantage o f information asymmetry. Such a database would also enable the insurance market to offer more efficient pricing for motor insurance covers to consumers. Integrated claims databases are not uncommon indeveloped countries. - 51 - 3.23. Changing UnderwritingRequirements and a Poolfor "Bad" Drivers. Given the mandatory nature o f the TPL motor coverage, the existence o fminimum tariff, and the political pressures on state owned companies to provide insurance coverage at loss making rates, insurers have neither the roomnor the incentives to improve the quality o ftheir underwriting. Yet, this has to change ifcompanies were to continue inthe business. A more selective underwriting however, currently practicedby private companies, i s likely to leave many drivershehicle owners with a poor driving record, without insurance coverage which i s likely to be socially andpolitically unacceptable. A potential solution lies inthe creationo f specialized government sponsored motor insurance pools for drivers with "inferior" driving record. 3.24. Establishing an Insurance Mutuat5for the Motor Industry. Inthe current information vacuum, there seems to be a role for a specialized insurance mutual owned and operated by truck ownerdfleet operators/drivers. Dueto a better alignment o f incentives o fthose insured and mutual policyholders insuch an institution, it would be well positionedto control moral hazard andadverse selection problems that currently plague this segment o f the market. 3.25. Our recommendationto the IRDA wouldbe to look into the possibility o flowering initial capital requirementsfor a such a mutual insurer given that it would offer coverage only to the members o f the industry (say, commercial motor insurance and related coverages only) and thus will not pose any threat o f insolvency to a broader public. This change however should be made contingent upon the ability o fmutuals to meet minimuminsurance operations requirements. Legal Framework and Courts 3.26. Statute oflimitations. Inthe absence o f statute o f limitations insurance companies are faced with an increasing threat o f fraud and overall uncertainty with regardto the reserving practices. It is therefore proposedthat a legal act sponsored by the IRDA should be prepared to amend the MVA with a view to introducing the Statute o f Limitations on the claims' filing rights o faccident victims. A short butreasonableperiod (say 3 months) should be considered. 3.27. UnlimitedLiability. While the industryappears to be mixed about the real harmdone to insurers' balance sheets bythe provision o f unlimitedliabilityinthe MVA, there i s certainly a strong case for introducing statutory limits on the insurers' liabilityto reduce financial uncertainty and consequently, ina fully liberalized market regime, achieve efficient pricing o f insurance products for consumers. It i s thus recommendedthat an amendment similar to that of the Statute o f Limitations bepreparedand introduced to the legislatureby the Insurance Regulator. 3.28. Motor Claims Tribunals. There is strong evidence that due to the growing volume o f cases the claims tribunals are overwhelmed, which results inprotracted waiting periods that on average take between 3-6 years. Facingthe lack o ftrained staff, office space and equipment, the claims tribunals call for a major overhaul which cannot be achieved without considerable changes inthe system o f its financing. Currently, the tribunals are funded by 45 Ina mutual insurance company policyholders are also ownersof the company, which means that the risk and rewards are completely mutualized. - 52 - annual budgetary allocations which do not reflect the growing annual volume o f claims. It is therefore suggested that a system o f court fees be introduced to enable the tribunals to recover the administrative costs from the claimants and thus boost its own administrative capabilities to process claims. Such fees would also serve as a deterrent to claims without merit or smaller claims and would help reduce the waiting time. 3.29. Claims Thresholds. Introducing thresholds, e.g., the levels at which an injuredperson can make a tort liability claim, may also significantly reduce the claim count and thus the backlog of claims inmotor claims tribunals. Such an approach is the most common form o f no-fault legislation, as it partially restricts the right to sue but does not completely eliminate it. Bodily injuryclaims that do not cross the threshold are limited to payment o factual economic loss -that is, medical expenses, a percentage o f lost income, and substitute services expenses. When the threshold limit is exceeded, the injured person, inaddition to his or her claim for actual economic loss, can make a claim for non-economic loss factors against the at-fault motorist. - 53 - Iv.KEYRECOMMENDATIONS AND THEIRIMPLEMENTATION Future development o fthe Bank's work inthe road sector inIndia i s likely to be extensively influenced by the findings and recommendations o fthis report. The findings presented on the trucking industry, on inter-city bus services and motor insurance have important implications for the benefits and costs o froad sector investments. For instance, the present long delays to trucks at State border crossings tend to nullify the benefits from higher speeds and reduced congestion en route. Consequently, addressing border crossing delays mustbeapart of our roadsector development strategy inorder to obtainthe fullbenefits o f the investmentsundertaken. Inthe following we summarize the main recommendations o f the report and indicate how these could betranslated into programs and actions under various transport sector operations financed bythe Bank.It should be emphasized that while the Bank can play a knowledge-sharingand advocatingrole in formulating these recommended reform initiatives, their effective implementation ultimately depends on the commitment and policy action by relevant federal and state government authorities. Trucking IndustryRecommendations 4.1. The report recommends consideration o f a system such as the EuropeanT.I.R. to reduce delays at state border crossing checkpoints, particularly for highvalue or time-sensitive goods. Such a systemwould permit sealed trucks, which elect to use the system, to operate without en- route inspections on the basis o f a certificate issued at origin by a duly authorized andbonded issuing entity. Since the present systemo f checkpoints i s administered by the States, and involve at least four agencies (sales tax, excise, motor vehicles and forests), changes would need to be coordinated across agencies and states nationwide. Initiatives bythe Bank should therefore be pursued at the national level, for instance with the Committee o f State Ministers coordinating tax and trade issues inrelation to implementationo f the VAT. 4.2. Policies to encourage the use o fmulti-axle vehicles and tractor-trailer combination, would help reduce transport costs and roadpavement damage. Towards this end, incentives such as reduced tax rates and tolls favoring such vehicles could be introduced. The Bank would encourage MoSRTH to introduce such measures within future budgets and toll rate policies. 4.3. Since a significant portion o f the driver population i s illiterate, it i s recommendedthat audio-visual driver training materials be developed inthe local language. This i s already being done inthe ongoing Kerala and Kamataka state road projects, and could be hrther pursuedinthe new state roadprojects proposed for Bank financing.. 4.4. To prevent excessive hours o f driving, it is recommendedthat trucks operating outside their home state be requiredto carry two licensed drivers at all times. This too could be taken up as a policy initiative under various State roadprojects. 4.5. A number o fpolicy changes are recommendedto improve axle load controls: expand enforcement authority beyond officials o f the Motor Vehicles Department (for instance Kamataka State has empowered PWD engineers); distinguish betweenminor (up to 5%) and more excessive overloading for which there would be extreme penalties; and make abetment o f overloading an offence so as to enable action against the broker or transporter arranging the load. - 5 4 - Enforcement of axle load controls i s critical to sustainability of road investments, and future Bank loans inthe road sector should support effective enforcement o f axle load control. 4.6. Invest inpermanent weigh stations at strategic locations on the National Highway network to enable random checks o ftrucks passing the weigh station whenever the station i s open. Require trucks found to be over-loaded to unloadthe excess load at their own cost and risk. This could be supported inthe next nationalhighwaysproject financed by the Bank. Recommendationson Inter-City Bus Services 4.7. The STUreform proposals advancedbythe ASRTU are unlikely to produce the desired improvements ininter-city bus services or stem the losses incurredbythe STUs, and consequently, it i s recommended that the strategy for STU reforms be reviewed. It also needs to be noted that STUs now constitute a declining share o fthe market for inter-city bus services and consequently the thrust o fpolicy inthis area should be on ensuring the health and competitiveness o fthe industry as a whole rather thanjust the STUs. 4.8. Reforms inthe Inter-City Bus services sector should include i.a. deregulation oftariffs, restructuring and commercialization o f STUs, elimination o f STUmonopoly rights, changes in the tax regime to achieve uniformity o ftax treatment o f all buses operating inthe inter-city markets, and creation o f an independent agency to establish, monitor and enforce competition rules, ensure access to common user infrastructure (terminals, bus stops), and last but not least, the improvement and enforcement o f safety regulations inrespect o f driver training, vehicle design and condition, and operatingprocedures, e.g. driver working hours. 4.9. The appropriate public policy for the inter-city bus services wouldbe to remove quantitative regulationsrestricting entryinto the inter-city bus transport markets, andto allow marketforces to determine bothtariffs andthe types ofservices offered. These proposals could be followed upwithin the Bank'songoing dialogue with MoSRTH. Recommendations on the Motor Insurance Industry 4.10. Itis recommended that switchingto a system where experience-rated premiumsattach to the owner and the driver, not to the vehicle, be taken up as a matter o f highpriority by IRDA. This will also require the development o fanintegratedclaims database andthe Bankcould consider supporting this as part of its technical assistanceprogram. 4.11. IRDA should also explore the creation o f a motor insurance pool for bad drivers who have been deniedcover by the insurance industry. 4.13 Finally, the report recommends amendment o f the Motor Vehicle Act o f 1988 to remedydeficiencies with respect to motor insurance such as the lack ofprovisions regarding a statute o f limitations, liability limits and thresholds for claims adjudication. 4.14 IRDAis already pursuinga number o fthese reforms, andthe Bankcould work with MoSRTH and IRDA on preparingrevisions to the Motor Vehicles Act neededto address the motor insurance industryissues. - 55 - A NN EX E S ANNEX1:ECOKOMIC OPERATING COSTSFOR DIFFERENT TRUCKS INDIA IN 1. The financial costs data o f Table 4 o f the maintext have been converted into economic costs usingthe conversion factors shown inTable A1.1 below. Economic prices have been derived fi-om financial prices by simply excluding taxes, inaccordance with current practice inIndia. This simple approach does not take into account such factors as the shadow price of labor and foreign exchange. However for the purposes o fthis study, such a simple approach i s appropriate and i s consistent with the approach adopted by the Central RoadResearch Institute (CRRI) intheir update o froaduser costs. Note this is a long-run, not short-run analysis; inthe very short-run, ifthere i s a surplus o fvehicles, the economic scarcity value o fvehicle waiting time would be zero. The long-run is the appropriate time frame for policy development designedto improve the efficiency o fthe industryover time. 2. Tax rates vary considerably between states and so only approximate conversion factors can be estimated, however they give a reasonable indication o f economic prices. For example the resulting factors are similar to those used inthe CRRI study mentioned above andproduce economic unitprices that are similar to those assumed inrecent economic evaluations ofroadprojects46. Table Al.l. EconomicConversionFactors Item Conversion Factor Fuel 0.50 Lubricants 0.77 I Tvres I 0.77 1 Spares 0.77 Crew 1.oo Maintenance LabodRepairs 1.00 Wayside Expenses 0.30 Overheads - - Staff and Administration 1.oo - Tax 0.00 I - Interest I 1.65 I - Depreciation 0.77 - Other 0.90 I - Profit I 1.00 I NOTES: (1) The conversion factor for fuel assumes an economic fuel price o f USD0.20 per litre. (2) The conversion factor for lubricants, tyres, spares and vehicles (depreciation) assumes an average combined tax o f 30% (mainly excise and sales taxes). (3) Staff and administration, profits and maintenance labor are regarded as mainly labor. (4) Wayside expenses are assumed to be 70% bribes (for which a conversion factor o f 0.00 i s appropriate for such transfers) and 30% administration costs (conversion factor o f 1.00) in accordance with the findings o f Trucking Operations in India, AITD, November 1999. (5) Conversion factor for interest i s based on assuming an economic opportunity cost o f capital o f 12%averaged over a ten year vehicle life. SOURCE: Consultant's estimates based on figures in Central Road Research Institute, Updation ofRoad User Cost Data, (2001). 46"HDM4 Representative RoadUser Costs for India", WorldBankNote, RArchondo-Callao, June 2000. - 56 - Cost per truck km 6.95 1 7.15 I 9.57 I 15.41 Cost per ton kmof capacity 1.39 I 0.79 I 0.60 I 0.57 4. To provide a basis for applying these cost estimates to different situations, the main assumptions about unit prices and utilization rates are summarized inTable Al.3. 5. Note that the assumed average load factor is about 100%. This reflects current operatingpractice where few trucks operate without load (especially over long distances) and there i s frequent overloading -typically 30-40% o ftrucks are overloadedbybetween 25% and 50% according to some operators. Of course trucks with specialized bodies such as tankers only achieve load factors o f about 50%. The reluctance to operate without loadwould be one reason for the low utilization o f trucks (only 80,000 kmper year, although higher rates are achieved with two drivers operating shifts). 6. The economic vehicle operating costs giveninTable Al.2 for the 27 ton tractor- trailer was based on an assumption that the fleet o f such vehicles consists of one trailer paired with one tractor unit. The vehicle operating costs for tractors infleets with alternativeratios of trailers/tractors has been estimated inTable Al.4 based on the figures inTable Al.2. The figures for a trailerhractor ratio o f 1.O are the same as those inTable Al.2. For higher values of this ratio, the costs have been increased assuming that: 0 utilisation o f the tractor increases inproportion to the carrying capacity of the trailer fleet, from 80,000 km to up to 160,000 km for a tractor in a fleet with a trailedtractor ratio o f two - 57 - variable costs such as fuel, tyres, spares and crew increase in proportion to tractor utilisation Table A1.3 MainAssumptionsfor TypicalTruck OperatingCostEstimates Item Assumption Vehicle - Carrying Capacity 9 ton - financial cost -- economic cost Rs. 800,000 (USD 16,000) Rs. 615,000 (USD 12,300) lifetime 800,000 kmover aboutten years - annual utilisation - operatinghoursper year 80,000 kmper year 2,200 - AverageLoadFactor About 100% Diesel Fuel - Financial Price Rs. 18 - 24per litre (USD0.36 -0.48) dependingon state - Economic Price Rs. 10.0per litre (USD0.20) - ConsumptionRate 4.5 kmper litre Tyres - Unit Financial Price Rs. 11,000 for a newtyre (USD220) - UnitEconomic Price - Lifetime Rs. 8,500 (USD 170) 70,000 km(including one retread) I - Crew Der Truck I 1.5 drivers and 1.0 heher I - MaintenanceStaff per 0.25 Truck - Average Monthly Earnings Rs. 3,000 (USD 60) including expense allowances 7. The increase inutilization assumes that the schedulingo f tractor andtrailer movements can maintainpresent levels o ftrailer utilization -this i s reasonable under suitable circumstances provided operating practices are changed to allow more intensive operation o f the tractor, including allocating tractors to more than one driver and operating more thanone shift per day. 8. Truck operators inIndia report that the ratio o ftrailerhractor prices i s higher than assumed above, but this seems unlikelyto apply infuture because, based on international experience, the ratio i s usually lower than 50%. Therefore the assumed increase incapital cost i s a compromise estimate that would apply inthe next few years. The assumed tractor and trailer prices apply to currently available vehicles that have beentraditionally manufactured inIndia. Recently more modem tractors and trailers have beenintroduced into the country, costing significantly more. 9. lo%, As showninTable Al.4, the operatingcost o ftractor-trailers canbe reduced by about from Rs.15.4 to 14.0 per km, through increasingthe number o f trailers operated by tractors. Inother words, the increased capital cost o f acquiring more trailers i s more than outweighed bythe advantages o f lower costs offered by increased tractor utilization. - 58 - Table Al.4. Annual Economic Tractor-Trailer Vehicle Operating Cost (Rs. per truck) - Profit 64,200 69,550 74,900 85,600 Total Overheads 442,458 496,190 549,921 657,384 Total Cost 1,232,656 1,483,937 1,735,218 2,237,780 Annual Utilization(km) 80,000 100,000 120,000 160,000 Cost per truck km 15.41 14.84 14.46 13.99 Cost Der ton kmof caDacitv 0.57 0.55 0.54 0.52 NOTE: For a tractor-trailer with 27 ton carrying capacity. SOURCE: Consultant's estimate - 59 - ANNEX2: COSTSDUETOADMINISTRATIVECHECKSAND FACILITATION PAYMENTS Traffic Flow 1. As reviewed inthe roaddevelopmentplanvision: 2021 by the IndianRoads Congress and the Ministryo fTransport (November 2000), various estimates have been made o f traffic flow inthe past, somebased on more reliable and comprehensive data than others. The approach adopted inthis work i s based on the sources deemed most reliable -namelythe national traffic statistics, complemented by detailed traffic surveys at particular locations (usually carried out as part o f feasibility studies for road rehabilitation projects). 2. The roadnetwork o f India includes 57,700 kmo fnational highwayplusabout 124,000 kmo f state roads. Inaddition there is a vast network of localroads. The best estimates o f truck delays are available from surveys on the 13,252 kmo fmainhighways that form the GoldenQuadrilateral and the mainnorth-south and east-west roads o f India. The estimate oftruck delays throughout the whole o f Indiais therefore based first on estimating the delays recorded or reported on these mainhighways and then applying appropriate factors to allow for delays experienced on other roads. 3. The average characteristics o ftraffic flow on these 13,252 kmofmainroads is estimated inTable A2.1 from national traffic statistics andrecent classified counts, especially those carried out as part o f feasibility studies for the World Bank's highway projects andwhich were reviewedespecially for the current work, as described inAllison Davis (2002), "Data Collection for Efficiency o f Indian Road Transport Study." Table A2.1 EstimatedAverage Traffic Flow on the MainNationalHighways (2002) NOTE: Typical average traffic composition on the Golden Quadrilateral and main East-West andNorth-South Roads. SOURCE: Consultant's estimate based on national traffic statistics, Davis (2002), and classified counts described in RoadMaintenance and Corridor Management for National Highways System inIndia (2000), Feasibility Report Consultancy Package VI for World BankNational Highways Project (1999), and CRRI, Updation o f Road User Cost Data (2001). - 60 - 4. This estimate of overall traffic level on this core part ofthe roadnetwork, with the highest flows, is consistent with recent National Highway Statistics on 30,000 kmof national highwayswhich have an average ADT o f about 23,000 PCU. It i s also consistent with the average ADT for motorizedvehicles of about 10,000 vehicles (minusmotor bikes) estimated for 25 typical road sections of the main highway network, o f total length 992 km, investigatedby Davis (2002). There is no reliable estimate of the overall proportions o f vehicles of each type on the whole o fthe main roadnetwork, but the values assumed inTable A2.1 are consistent with available overall estimates ofpassenger andfreight traffic such as those inthe RoadDevelopment Plan Vision: 2021 (2000). Truck Delays 5. The greatest uncertainty inestimating the economic impact o f truck delays i s due to the lack o f information about the incidence, lengthandnature of the delays. Only one formal survey has beencarried out inrecent years, and this covered only three truck jo~rneys.'~ The results from this surveyhave therefore been cross-checked with other, more informal estimates. 6. Inthe formal survey, thetime spent at check points onthe mainnational highways was measured by traveling observers on three typical long distance trips. The results are summarized inTable A2.2. Delays were recorded at bothinter-state andintra- state check points and duringthe preparatoryperiod at the start o fthe trip. The inter-state check points included checks o f inter-state transport permits (Road Transport Officials) andsalestax documentation. The intrastate check points were mainlydue to local octroi tax collection activities. Inaddition there were random checks bypolice and RTO officials who check transport documents and inspect vehicles. TableA2.2 RecordedJourney Time ofTrucks on NationalHighways SOURCE: SundarCommitteeReport, Annexure 3.1 (November 1999). 7. Inanother, lessformal surveycarriedout onanotherroutein2002, between Hyderabadand Chennai, an observer traveling on a truck has noted that about 2 hours were spent at 12 check posts over a distance o f 625 km. This is equivalent to 0.0030 hoursper km.All delays were modest induration andwere confinedto formal check points. Time was spent gettingthe first control stamp before the trip beganandno random checks bypolice or RTO officials took place. 47Bythe CentralInstitute of RoadTransportinJune-July 1998, for the Sundar Committee. - 61 - 8. Operators report that delays at check points can vary considerably, fi-om a few minutes inmost cases to delays of several hours, or even days ifthere is some problem with documentation. Some operators claim that trucks spend a total time o fup to two or three hours at check points per day, equivalent to 5-10 hours per long distance trip, although this could not be confirmed. 9. Itis clear that the occurrence o fdelay varies considerably betweenone state to another because the sales tax controls at inter-state borders differ inaccordance with the prevailing sales tax rates inadjoining states. Furthermorealthough no such difficulties were observed on the four trips described above it is clear that some longdelays do occur from time to time andthis would add significantly to the overall average. For example it was observed duringthe survey that the particular drivers being observed knew the likely locations o fmobile enforcement officers and had already established close relations with the officials on their routes. Other drivers with less experience o fparticular routes would belikely to experience longer delays. 10. Ittherefore seems reasonable to assume that the range o f delay time experienced at check points on the mainhighways i s usually between 0.0015 and 0.0040 h o u r s h . The associated delay with gettingthe paperwork sorted out i s similarly estimated to be between 1.6 and 3.0 hours per trip (equivalent to between 0.0010 and0.0020 per kmfor typical long distance trips). User Costs 11. Two types o f user costs are affected by truck delays: 0 vehicle operating costs that are fixed rather than vary with distance operated, 0 cargo time-related costs, which are mainly cargo holding time costs caused byincreased inventories. 12. There are two main issues inestimating these user costs: fixed v/s variable costs and the extent to which time savings can be usedinpractice. (a) Fixed Versus Variable Costs 13. This issue is about the extent to which costs are fixed rather than vary with distance: delays do not affect costs that vary only with distance, but they may increase costs that are fixed because an increase injourney time would tend to increase the numbero ftrucks requiredto meet demand. Non-overhead vehicle operating cost elements such as fuel, tyres, spares and maintenance labor are likely to vary almost entirely with distance traveled, so they can generally be excluded from the assessment o f delay time cost. Onthe other hand, overheads such as interest payments (or, ineconomic terms, the opportunity cost o f capital) can generally be regarded as fixed, and therefore would contribute to delay cost. However the situation i s not so clear inthe case o f crew costs, administrative costs and depreciation (which may vary with distance to a greater or - 62 - lesser extent ,in other words, some o f these costs are likely to be lower for vehicles which have lower utilization). 14. This study makes the simple assumption that non-overhead cost items such as fuel, lubricants, tyres, spares, crew, maintenance labor andwayside expenses all vary wholly with distance traveled. For overhead cost items, interest, depreciation and profit are regarded as fixed whereas other overhead items (which includebroker fees and motor insurance fees) are regarded as variable. This should result ina reasonable estimate o f fixed costs because effects of delays such as additional fuel consumed by idling vehicles at check points (assumed to be variable but inpractice somewhat fixed) will be offset by effects such as reduceddepreciation causedby lower vehicle utilization (assumed to be fixed but inpractice somewhat variable). 15. The main contributor to cargo time costs is the cargo holdingtime cost and so the study considers this to be a wholly fixed cost, varying with time rather than distance. (b) Extent to Which Time Costs can be Realized in Practice 16. Small savings intrip time maybe unlikelyto achieve significant benefits interms of reducedvehicle fleet andgeneral overheads, because there i s little scope for alternative ways o f exploiting these resources. However inthe case o f delays at check points, significant time savings can be generated duringthe course o f a longjourney, amounting to several hours of truck and staff time. Since this creates opportunities for rescheduling vehicles and staff, real cost savings can be expected from reducing delays at check points. 17. For the same reason, cargo holding time costs can be expected to achieve significant inventory cost savings, although o f course this varies with the value ofthe commodity. 18. The annual fixed costs for eachtype o ftruck have beenexpressed inTable A2.3 interms ofRs.perhourbydividingthe annualoverheads bythenumber ofoperating hours per year (2,200 hours for most heavytrucks, representing the typical current operatingpattern o f trucks inIndia4*). The cargo holding cost for each truck type has also been estimated inTable A2.3 based on the measured average value inRsper ton o f commodities carriedbytrucks4' and the average load carried by each type o f truck. 19. The impact on the mainhighways is based on the flows estimated inTable A2.1 above, while the estimate for other highways is roughly estimated assuming that they have a daily flow o f about 13,000 pcu comparedto about 26,000 pcu on the main highways. This i s a reasonable assumption considering that state roads, which are next lowest in the roadnetwork hierarchyhave a recorded average daily flow o f about 5- 10,000 pcu. 48"HDM4 Representative RoadUser Costs for India," World BankNote, RArchondo-Callao, June 2000. 49"Updation o f RoadUser Cost Data," Central RoadResearch Institute, Delhi, July 2001. - 63 - 20. Delays (per km) are assumed to be 50% lower on roads that are not on the main highway network. This i s reasonable considering that the trucks operating on these other roads would be operating on shorter distance hauls which are less likely to cross administrative boundaries, and may therefore be less subject to tax and permit controls. Delays that may occur on state roads are not included inTable 7 o fthe main text so the estimate o f overall delays inIndia shouldbe considered to be a conservative estimate. UnofficialFacilitation Payments 21. Unofficialpayments such as bribes and donations made by drivers inthe course o f long distance trips have been recorded bytraveling observers on the same three trips on which the truck delays were recorded. The results are described inTable A2.4. Unofficialpayments includedbribesto enforcement officials, donations exacted by villagers for festivals and fairs, and payments to staff at loadinglunloadingpoints. 22. Most o fthe payments were made at check points (RTO, tax inspectors or Octroi officials) or duringother road-side checks bypolice or RTO officials. These payments were particularly highon the trip betweenMumbai and Delhibecause the truck was overloaded by 3-4 tons. 23. Duringsimilar observations in2002 on a 625 kmtrip between Hyderabadand Chennai, total unofficial payments amounted to Rs.30 (to government officials). The low level o fpayment probably reflected the fact that the truck did not encounter any road-side police or RTO checks, which normally are reported to incur bribes o f about Rs. 20 per inspection. Some truck operators claim that up to 10%o frevenue is paid inbribes, which would be equivalent to Rs.1,500 per long distance trip, which i s higherthan recordedin the surveys. Other truck operators report that typically at least Rs. 50 per day is paidin bribes (Rs.200 pertrip). Itis understood that some truck owners make regular payments to RTO officials to enable free passage o ftheir vehicles without check point delays or bribes-this couldbeone reason for low levels o fbribes paid enroute by some operators. 24. Based on these observations it seems that unofficial payments, including those not paid en route, vary widely but usually range between Rs.200 and 1,500 per trip (Rs. 0.13 and 1.OO per km). The bribes are reportedly normally higher for overloadedtrucks. - 64 - Table A2.3 Time-related Road User Costs Item Type of Truck Light Medium Heavy Multiaxle (A) Vehicle OperatingCost - annual overheads (Rsltruck) 89,790 129,624 170,812 341,298 - Annual hours per truck 1,800 2,200 2,200 2,200 - Cost pertruck hour (Rs) 49.9 58.9 77.6 155.1 (B)Cargo HoldingCost - Average loadper truck (ton) 5 9 16 27 - Average value (Rs./ton) 26,107 26,107 26,107 26,107 - Cost per truck hour (Rs) 8.7 12.8 22.8 38.4 (C) TOTAL COST (Rshour) 58.6 I 71.7 I 100.4 1 193.6 NOTES: (1) The vehicle operating costs are basedon annual overheadsper truck estimated inTable 5 inthe informal technical note on methodology (excluding "other" overheads). (2) The averagecargo value is basedon a measuredfinancial value of Rs 26,667 per ton recorded for long distance trucking in2000, increasedby 10% to apply to 2002. The economic cargo holding cost is estimatedassuming a conversion factor of 0.89 (equivalent to an average tax of 12"/) and a 12%opportunity cost of capital. SOURCE: Consultant's estimate Table A2.4 RecordedEnRoute Payments by Trucks on National Highways Route Distance Total UnofficialPayments at Other Unofficial Expenses Check Points Payments (e.g. (including other road- loadinghnloading) side police and RTO checks) Km Rs Rs R s h Rs R s h Mumbai -Delhi 1,430 8,100 1,880 1.31 100 0.07 Delhi-Kolkata 1,490 6,550 650 0.44 0 0.00 Kolkata-Chennai 1,845 7,500 561 0.30 123 0.07 Average 1,588 7,383 1,030 0.65 74 0.05 SOURCE: Trucking Operationsin India, AITD, November 1999 - 65 - ANNEX3: PROPOSEDAMENDMENTS THE MOTOR TO VEHICLE ACT1988TO OVERCOME OVERLOADING I Present Provision Proposed Amendment Section 114(1) Section 114(1) Any officer o fthe Motor Vehicles Department Any officer authorized inthis behalfby the State authorised inthis behalfby the State Government Government shall, ifhe has reason to believe that shall, if he has reasonto believe that a goods vehicle a goods vehicle or trailer i s being used in or trailer is being used incontravention o f contravention o f Section 113, may require the Section 113 requires the driver to convey the driver to produce a certificate or other proof from vehicle to a weighing device, ifany, within a a government authorized weighingdevice and in distance of ten kilometres from any point on the case o f failure to produce such a certificate, forward route or within a distance o f twenty require the driver to convey the vehicle to a kilometres from the destination o f the vehicle for weighing device, if any, within a distance o f ten weighment; and if on such weighment the vehicle is kilometers from any point on the forward route or found to contravene inany respect the provisions o f within a distance o ftwenty kilometers fromthe Section 113 regarding weight, he may, by order in destination o f the vehicle for weighment; writing, direct the driver to off-load the excess (2) (a) Ifthe laden weight i s found to be within weight at his own risk and not to remove the vehcle the gross vehicle weight specified inthe or trailer from that place untilthe laden weight has certificate o f registration o f the vehicle, the been reduced or the vehicle or trailer has otherwise vehicle shall be allowed to proceed. been dealt with so that it complies with Section 113 (b) Incase the laden weight exceeds the gross and on receipt o f such notice, the dnver shall vehicle weight specified inthe certificate o f comply with such directions. registration o f the vehicle by upto 5%, the fines specified inSection 194 shall be imposed by the authorized officer and the vehicle allowed to proceed to destination. (c) Incase the ladenweight exceeds the gross vehicle weight specified inthe certificate o f registration o f the vehicle by more than 5%, the authorized officer may, by order inwriting, direct the dnver to off-load the excess weight at h i s own cost and riskand not to remove the vehicle or the trailer from that place untilthe ladenweight has beenreduced or the vehicle or trailer has otherwise been dealt with so that it complies with Section 113. (3) Whoever drives a motor vehicle or causes or allows a motor vehicle to be driven in contravention o f Section 113 or Sub-Clauses (b) and (c) o f this Section shall be punishable with a minimumfine o ftwo thousandrupees and an additional amount o f one thousand rupees per ton o f excess load or part thereof. Section 114(2) Section I14(2) Where the person authorized under sub-section (1) The existing provision may be dropped. makes the said order in writing, he shall also endorse the relevant details o f the overloading o n the goods carriage permit and also intimate the fact o f such endorsement to the authority which issued that permit. - 66 - Section 188 Section 188 Whoever abets the commission o f an offence under The existing provision may be enlarged to include Section 184 or Section 185 or Section 186 shall be overloading. The amendedprovisionwill be as punishable withpunishment provided for the follows: offence. "Whoever abets the commission o f an offence under Section 113, Section 114, Section 184, Section 185 or Section 186 shallbe punishable with a fine and a term injail as provided for the offence". Section 194(1) Section 194(1) Whoever dnves a motor vehicle or causes or allows Reference to Sections 113 and 114 may be a motor vehicle to be drivenincontravention o fthe deleted inview o f amendment to Section 114 provisions of Section 113 or Section 114 or Section suggestedabove. Section 194(1) shall read as 115 shall bepunishable with minimumfine o ftwo under: thousand rupees and an additional amount o f one Whoever drives a motor vehicle or causes or thousand rupees per ton o f excess load, together allows a motor vehicle to be dnven in with the liability to pay charges for off-loading o f contraventiono f Section 115 shallbe punishable the excess load. with minimumfine o f two thousand rupees and an additional amount o f one thousand rupees per ton o f excess load or part thereof, together with the liability to pay charges for off-loading ofthe excess load. Section 194(2) Section 194(2) Any driver o f a vehcle who refuses to stop and Add"or imprisonment upto one month" at the end submithisvehicle to weighmg after being directed o f the Section. The Section shall thus read as to do by an officer authorized inthis behalf under follows: Section 114 or removes or causes the removal o fthe "Any driver ofa vehicle who refuses to stop and load or part o f itprior to weighing shall be submithis vehicle to weighing onbeing directed punishable with fine which may extend to three to do by an officer authorized inthis behalf under thousand rupees. Section 114 or removes or causes the removal o f the load or part o f it prior to weighing shall be punishable with fine which may extendto three thousand rupees or imprisonment upto one month". Section 200(1) Section 200(1) Any offence whether committedbefore or after the Delete Section 194 from sub-section (1) o f commencement o fthis Act punishable under Section 200. Sections 177, 178, 179, 180, 181, 182, Sub-section (1) or (2) of Section 183, Sections 184, 186, (Section 189, Sub-section (2) or Section 190), Sections 191, 192, 194, 196 or Section 198, may either before or after the institution o f the prosecution, be compounded by such officers or authorities and for such amount as the State Government may, by notificationinthe Official Gazette, specify inthis behalf. - 67 - ANNEX4: CHINA-THEEFFICIENCY ROAD OF TRANSPORT INDUSTRY Backmound 1. The rapid economic development o f China has created hightransport demand since 1980. Passengertraffic grew by4.8 times to 1.32 trillionpkmin2001 from 228 billion pkmwhile the freight traffic grew at a rate o f 7 percent to 4.76 trillion tkm from 1.16 trillion tkm inthe same period. The roadnetwork nearly doubled over the same period, from about 0.9 millionkmto 1.76 million kmo fwhich 50 percent i s paved. China also embarked on the construction o fmajor arterial corridors to facilitate the movement of freight to support the growing foreign trade and industrialization. The additional road lengthmainlyconsisted of higher levelhighways and expressways thereby creating more additional capacity thanjust the increases inlengthwould indicate. The road network includes about 125,000 km o fnational highways o fwhich 25,000 km are expressways and another 27,500 km are divided 4-lane highways. .2. The numberof motorized vehicles (excluding agricultural vehicles andtractors) increased ten times to 18 million inthe same period. However, the number o fnon- motorized vehicles reduced duringthe same period. China has poor road safety records. In2002, itrecorded 770,000 roadaccidentsresultingin 109,000 fatalities. Itis estimated that almost 70 percent o f accidents involved trucks. InstitutionalArrangement 3. The HighwayAdministration Bureauo f Ministryo f Communication (MOC) is responsible for roadtransport operations, road construction andmaintenance. Its Planning Departmentplans the development ofthe nationalroadnetwork andformulates policies. Communication departments o fprovincial governments are responsible for road administration. Transport administration is with the Highway Transport Administration Bureauwhile traffic and safety is the responsibility o fthe Provincial Public Safety Bureau. 4. Initiallythe local communications departments owned road construction and freight and bus service companies. As the reforms deepened the role of the state shifted gradually from that o f owner, investor andmanager to that o fregulator and policymaker encouraging the development o f independent providers o f services. This ledto detaching o fmost o fthese State Owned Enterprises (SOEs) from the local communications departments. However, some localdepartments still own andoperate bus andtruck companies. Regulatory Framework 5. Passengerand freight services were initially operated as SOEs. However, in 1986, the State Economic Commission, now abolished, approved a tentative regulation on management o fhighway transport (Joint Regulation). This set out the regulatory 68 framework for licensing, customer and operator relationship and rules for provision o f transport services. It marked a significant milestone inthe transformation o f aplanned economy to a market-oriented economy. Since then M O C has issued a series o f regulations to align road transport services with the market economy and encourage private provision o f services and competition. 6. Licensing System:The Joint Regulation provides the basic framework for the licensing system. Licenses must be obtained from the appropriate transport authority (provincial, prefecture, or county communications departments). This regulation also addresses licensing criteria pertainingto social needs, the applicants' capability to provide services, business scope, technical qualifications, and operational conditions. Normally, the licensing criteria including the following: (a) size o f fleet owned bythe enterprise, (b) condition o f vehicles, (c) adequacy o fparking facilities, and (d) proper certification for technical staff incharge o f safety control and vehicle maintenance. To facilitate the entry o f individuals, it has, however, set out more lenient criteria pertaining to number o f vehicles, their condition, parkingand compliance with related regulations. 7. A second licensing system is vehicle permits, which ensures every bus andtruck used for service meets safety and economic criteria. It is designed to check engine performance, the operation o fmain drivingcomponents, and the level o fpollutant emissions. Vehicle permits must berenewed annually, are valid throughout China and must be carried onboard the vehicle. 8. Price Control: For decades the policy has been to allow local governments to set prices under general guidelines issuedby MOC. Proposedtariffs are submitted to the pricing authorities o f each local government for approval. Once approved, tariffs are reportedto MOC. To ensure observance o fthese tariffs and monitor their application, transport operators are required to use tariff vouchers on which the basic rules o ftariff applications are printed and submit these vouchers to appropriate government authorities. 9. As apart ofthe economic reform, inthe last 15 years, policies have shiftedaway from rigidtariff control mechanisms. In1984, MOC issued a Regulation on Passenger and Freight Transport Tariffs (RPFTT), which permitted local governments to set cargo rates that differ from national standards. The RPFTT allowed for prices to vary by 20 percent inconsideration o f local conditions. Further in 1987, the State Planning Commission and M O Cjointly issued a Regulation on Motor Vehicle Transport Tariff Management, which left the power o f settingtariffs to the Provincial Communications Departments (PCDs) inconsultation with provincial pricing administrations. Today, the tariffs for passenger transport are still under the government control. Upwardadjustment can be made during the peak seasonwith prior public hearing. However, competition has resulted inmarket prices generally lower than the government publishedprices. Thus such publishedprices serve only as the ceiling duringpeak periods or highdemand routes. Infreight transport, fierce competition ledto the de-facto abolition o ftariff controls. 69 10. Check Points and Forced Stops: Ingeneral, passenger and cargo traffic moves freely throughout China. There are no check-points at provincial boundaries. The national law prohibits stoppingo f vehicles by any institution other thantraffic police. No fine can be collected inthe absence of traffic police. Inaddition, commercial taxes such as sales andcommodity taxes are collected where thebusiness is registeredrather than en-route. Fines for overloading have to be issuedjointly byroad authorities and traffic police. Road Transport System 11. Passenger: The interurbanbus service plays a vital role inmoving China's rural population and urbanlongdistance travelers. By 2001, there were 205,941 bus service providers throughout China. These operators ran 139,000 bus lines covering 99.3 percent o f China's communes andtownships and 91.8 percent o f villages. There are over 10,120 busterminals throughout Chinawith daily departure o f 1.04millionbuses. The interurban bus service expanded rapidly inthe past decade. From 1990 to 2001, bus transport output increased from 6.5 billion passengersto 14.0 billionpassengers, which is an average growthrate o f 7.3 percentper annum. Interms o fpassenger-kilometers, output increased from 262 billionto 720.7 billion(9.6 percent per annum). Bustransport share inthe inter-city trips increased from 46.6 percent in 1990to 54.8 percent in2001. 12. Passenger services are dividedinto four categories -inter-provincial, inter- prefecture, inter-county and intra-county and classified into 5 classes based on fleet size, assets, annual revenue, output and experience. The Class 1operator i s allowed to operate buses inany of the categories while the Class 5 operator can operate only intra-county services. 13. Tari$s: Tariffs vary significantly with the lines andregions the bus serves, the conditions o f the bus and the tolls the operator pays. InHebei Province, for example, the tarifffor regular buses (large bus) is about US 2-4 cents perpassenger-kilometer. Onthe other hand, inHubeiProvince, tariff i s about 2-5 cents per passenger-kilometer. This figure i s about 2 to 3 times o f rail passenger tariff rate. 14. Industry structure: The industryhas undergone a major change with many private operators enteringthe bus transport industry.However, the SOE still play the dominant role controlling about 55 percent o f total passenger seats available. SOEs generally have larger fleets with an average o f 21 buses while the private operators own on an average only 1.7 buses. This clearly indicates that the private operations are fragmentedwith individuals providing the services. Most private operators go for mediumand small size buses to provide intra-county and inter-county services. There has beena decline inthe fleet o f largebuses since 1996. 15. Freight: For many decades up until early 1990s, China's road freight service sector hadplayeda role subsidiaryto rail inthe nation's transport system. Duringthis time, the rail system has beenthe logistics backbone o fthe Chinese economy and the primarybasis for the nation's materialdistribution system. Itisparticularly well suitedto support a plannedeconomy. Evenas late as 1990, the highways' market share for freight 70 was only 24 percent of combined road andrail haulage. One ofthe mainreasons was the underdevelopmento fhighwayinfrastructure. Significantly, China's highwaymarket share for freight i s lower than for other countries at comparable phase o f economic development. As the shift towarda market economy accelerated, structural changes in Chinese industry have gradually shiftedthe productionmix to highervalue and lower density products, typically produced in smaller lot sizes. This shifihas increased demand for faster, morereliable and flexible transportation services, i.e. trucking. 16. In2001, China's truckingindustrymoved 10.6billiontonsofcargowithatotal output 633 billion tkm. The annual average growth rates for cargo movement (intons) andhaulage (intkms) was 3.9 and 5.9 percent respectively. Highwaymarket sharehas grown from 24 percent to 30 percent o f combined road and railhaulage. 17. The freight services are dividedinto 6 categories based onthe type of cargo moved- normal cargo, less than truck load, oversized, container, refrigerated cargo, and hazardous material and firms are classified into 5 classes on the basis o f total capacity, annual turnover, assets and years o f experience. Class 1and 2 operators are allowed to operate all categories o f cargo while Class 5 could carry only normal cargo. 18. Industry structure: The trucking industry inChina is currently undergoing dramatic changes inthe number o f operators, composition andmarket share. The for-hire truckingindustryis made up oftwo distinct segments, (a) state-owned enterprises (SOEs) and (b) private operators. By2000, each o fthe two segments owned around 1million and 2.1 million trucks, respectively. Inaddition, there are 3.2 million own-account trucks. The private segment has grown at 16.2 percent per annumbetween 1996 and 2000. This stands insharp contrast to zero growth for SOEs duringthis period. The fast increase o f private operators is mainly dueto very few entry barriers inthe market. Virtually anybody who purchases a vehicle can obtain a license. Inmany cases licensing itself is not strictly enforced. This has now ledto over capacity, andthe nationwide average payloadcapacity ratio i s only about 0.5. 19. Fleet: Truck fleet has grownrapidly inrecent years (see TableA4.1). The annual growth rate o f for-hire andown-account trucks reached 8 and 5.9 percent, respectively. Correspondingly, the load capacity for each group grew 4.9 and7.2 percent. Table A4.1: Truck fleet SOURCE: Consultant's estimate. 71 Inthe for-hire truck fleet, heavytrucks grew at amoderate rateo f3.7% per annumwhile the mediumtruck actually declined. Onthe other hand, lighttrucks grew at a much faster pace than others. This indicates that the trucking fleet inChina is moving into two distinctive directions, one for lighter and the other for heavier trucks. The large trucking companies are replacing their old mediumsize truck with modern, efficient heavy trucks, while the farmers who mostly own a singletruck are going for smaller vehicles. 20. Tar@ Freight tariffs, eventhough set by the government, vary significantly dependingon operating conditions. For example, inAnhui Province, the tariffis US 4-5 cents per ton-kilometer while inZhejiang and Guangdong, the tariff i s U S 5-6 cents per ton-kilometer. On average, the freight tariff for roadtransport i s about 2 to 3 times that for rail transport. 21. Intermediary agencies: The trucking industryhas a number o f intermediaries who play important role infacilitating the road freight business. These include freight forwarders andbrokers. The freight forwarders usually are engaged inthe business o f collecting, forwarding or distributing goods carriedby trucks. Their revenues are from the difference between the tariffthey charge from the customer and the amount they pay to the truck ownedoperators. On the other hand, brokers take commissions from the truck as well as freight owner for providing demand and supply information. 22. Vehicle technology and overloading: About halfo fthe truck fleet inChina is made up o f small size trucks (withloading capacity less than 4 tons) and 10 percent o f mediumsize trucks. Both small andmedium size trucks are 2-axle vehicles andmost o f them are domestically produced. The heavy trucks are generally imported from Europe and USA, or jointly made inChina. Sales o fthese vehicles are rather limitedbecause o f highprices, eventhough thesemodels have muchhigher performance. 23. Since the RMF, HTMF and toll rates are based on designed loading capacity, to avoid taxes and tolls, carriers like to buy vehicles with low design capacity but higher carrying capacity. The local manufacturers tendto build vehicles meeting the customer's demand. The vehicles are built to carry additional load without adding extra axle, which results inoverloading, and i s very common throughout China. This causes a great deal o f damage to pavements and other structures. Realizing the severity o fthe damage, M O C and the Ministry o f Public Securityjointly issued regulations in2001 to prevent over- loading. The regulation calls for much stricter enforcement o f axle-load limits and heavier punishments. But the problem i s so widespread and opposition so strong that the newregulationis not consistently enforced. Some provinces are now pursuinga new approach bybuildingweigh stations at toll gates and are charging tolls based on actual weight rather thandesign capacity. Some are even experimenting with actual axle-load based toll rates to provide incentives for introduction o fmulti-axle trucks. 72 ANNEX5: PAKISTAN -THEROADTRANSPORT INDUSTRY Background 1. Pakistan's roadnetwork has grown many folds since 1947, from a mere 50,000 kmto over 250,000 kmbythe endo f2000. Duringthe sameperiodthe total numbero f motor vehicles has increased from 20,000 to over 5 million. With increased investment in the road sector compared to the rail sector insuccessive 5-Year Plans, the share o f road transporthas increased significantly. 2. Pakistan's roadnetwork has only about 9000 kmo fnationalhighways and motonvays runningnorth-southconnectingthe only major port, Karachi, to its hinterland and the major cities. The mainhighway corridor connecting the port also serves the land- locked countries o f Central Asia andAfghanistan. 3. Withincreasingdevelopment o fthe roadnetwork, roadtransport has become the major mode for boththe freight andpassengers. Roads now carry almost 90 percent o f the total freight whereas in 1947 it carried only about 15 percent. This has created strong pressures to develop better roads, and the Government o fPakistanhas planned the development o f access controlledmotonvays along the major transport corridors. From Lahore to Islamabadthe motonvay has already been inoperation andthe others are either underconstruction or are at the planningstage. InstitutionalFramework 4. The roadnetwork inthe country is the responsibility o fthe federal, provincial, district, local andmunicipal administrations dependingupon the type o f road. The network i s classified into motonvays/national highways, provincial roads, farm to market roads andurbanroads. The National HighwayAuthority (NHA), i s responsible for the motonvay / national highways, the Communications and Works Ministry of each province for provincial roads, andthe district administrations for the district and local roads. 5. Provinces are responsible for issuingpermits for freight transport operations and the trucks are free to operate all over the country on the particular route for which the permit is issued. The provinces have different fee structures for truck permits which affects the registration o fvehicles indifferent provinces. Most vehicles are registered,in the province o f Baluchistanwhich has the lowest charges. Price Control 6. There is no price control for the road freight and the freight haulage rates are determined by demand and supply. Inmost cases freight rates to Karachi are lower than from Karachi since most traffic originatedterminates at the port andthe amount o f outbound cargo from Karachi port is more than the inboundcargo. The only exception i s the N L C which was established in 1978 by the army's commercial organization. The prices for tankers are determined by the oil companies. 73 Road Freight Transport 7. Industry Structure: The trucking industry inPakistan i s mainly private sector. There are a large number of enterprises owning up to 5 vehicles, and a few with 20-25 or more vehicles. Inadditionto truck operators there are booking companies and freight agents. The booking companies book the cargo and get it shipped. The freight agents work as an intermediary between the shipper and the truck operator on commissionbasis. Small truck operators have formed an association to safeguard their interests, andprevent exploitation bybooking companies and freight agents. There were about 4000 such associations in 1994. 8. Fleet: The truck fleet, a total o f about 140,000 trucks, i s divided into three categories- 2-axle (66%), 3-axle rigid (9%) and 3 or multi-axle articulated (25%). The truck manufacturing industryi s relatively new inPakistan. Most o f the fleet (53 percent) consists o f 2-axled Bedford trucks manufacturedinthe country. The cost o f different trucks vary significantly, from a low o f about Rs.300,000 for a Bedford truck to 3,100,000 for an articulated Volvo truck with a carrying capacity o f 60 tons. Most o f the fleet i s old with an average age o f about 20 years. The multi-axle vehicles are generally newer. 9. Freight rates: The prevailingmarket structure of the road freight industry results inintense competitionforcing truck operators to quote prices attimes lower thantheir operating costs. The unit prices vary for different routes as well as byvehicle type. For a 2-axled truck carrying 12 tons the tariffranges from a low o f Rs.0.48 per tkm on Lahore- Karachi route to a higho fRs.l.04 per tkm on Lahore-Peshawar route. While for a multi- axle truck-trailer carrying about 50 tons, the tariffs are Rs.0.25 and Rs.0.77 per tkm for the respective routes. On the other hand, the unit operating cost, per tkm, of different vehicles have beenestimated at Rs.l.24 per tkm for a 2-axled truck and Rs.0.64 per tkm for a multi-axle truck based on their authorized loads. 10. Over loading: The over loading o f trucks is rampant inPakistan. The difference between operating costs andthe prevailing freight rates, as discussed above, necessitates over loading to recover even the operating costs. The legal limit for 2-axle truck is only 7 tons which is quite low by international standards. Therefore, most trucks are carrying almost double the legal axle loadwith an average o f 12 tons for a 2-axle truck. 11. Inthis context theNationalHighwayAuthority has embarked onaplanto establish weigh-bridges on the 8,479 km o f federal controlled roads inthe country. These weigh-bridges will be built at strategic locations, andwould check commercial vehicles as they enter the federal Highways and Motorways network. NHA is also coordinating with the provincial C & W Departments so that weigh bridges are installed onprovincial roads as well. Implementationwill be staggered over a period o f years. The highly effective National Highway and Motorway Police (NHMP) will enforce the program, in addition to their normal duties. There was a concerted effort to conduct an awareness campaign through the Transport Owners Associations so that bythe time the program became effective on lSt July 2003 the trucking industrywas ready to conform. The 74 customers o f the trucking industry will eventually bear the burdeno f the major changes inthe structure of overlandcamage, as itwill increase their transport costs. While amore efficient trucking industrymay eventually be realized inthe long term, difficult times are predicted inthe short and medium term. There is a need for concurrently raising the legal axle load limits. 12. Trafic Safety: Fromthe limitedstatistics maintainedbyvarious government agencies it has been ascertained that commercial vehicles inthe country are the main cause o f serious accidents on bothhighways and inurban centers. This includes all types of commercial vehicles (buses, trucks, tankers, taxis etc). While the numbers o f such commercial trucks are much less than the overall vehicle population, the percentage o f accidents where at least one commercial vehicle is involved is almost 70% o f the total. The factors that contribute to such apoor safety record are two folds -the machine and the driver. The age o fthe vehicle fleet has already been mentioned. By and largedrivers are illiterate and undisciplinedwith no incentive to improve their drivingskills. They are required to work under tough conditions -longworking hours that leads to fatigue, which i s a serious safety hazard. Yet another government program is under consideration to introduce reputable private driver training institutions that would also be involved in issuing drivinglicenses. This program has as yet not been implementedbut needs to be put inplace as early as possible. Goodroad engineering andbetter enforcement arethe two other effective methods o f improving safety on the network. Steps have alreadybeen taken to address boththese issues. The federal NHAandprovincial C & W Departments are upgrading the roads withintheirjurisdiction and have plans to install road hrniture that is presently lacking. Already positive results have emerged on the sections o fN5 (the National Highway) and M 2 (Motorway between Islamabad and Lahore) that have come underthe control of the federal NH&MP. Itis also expectedthat the restructured provincial police departments would exercise better control now that the Trafic Police havebeen separated from the main force. 13. Computerized Ehicle Registration: This has commenced inthe province o f Sindh closely followed by the provinces o f Punjab, NWFP and Baluchistan. It is logical that the program would record the various aspects o f each vehicle i.e. age, number o f axles, make, etc. but provisions for producingthe relevant statistical reports may not have beenmade yet. Eachprovince has developedtheir systems independent o f each other and it is likelythat their compatibility with one another has notbeentakeninto consideration. Itistherefore important to addressthis issue. This will bringmuch-needed efficiency in the enforcement o ftraffic rules inaddition to generating up to date statistics as and when required. 14. International Carriage of Goods by Road With the prospects o f overland transit trade through Pakistan between the CentralAsian Republics andthe seaports o f Karachi, Port Qasim and the under construction Gwadar port, there is a needto promote the movement o fvehicles and cargo over internationalborders. While all these land -locked countries includingAfghanistan have ratified the international Conventions covering TIR and CMR, Pakistanis the only country inthis regionthat has not yet done so. The TIR Convention allows the easy movement o f vehicles across international frontiers as it i s a 75 form o f a "passport" while the CMR Convention i s a Contract for the Carriage of Goods by Road and gives legal status to the haulier as an instrument o finternationaltrade. In view o fthe rapidly changing scenario inthis region, there seems to be little alternative for Pakistan but to ratify boththese Conventions. 76