ASSESSING THE EFFECTS OF TRUCKING REGULATION IN KOREA Prepared by: Graham Smith Transportation Department Tony Michell Hull University, England Buyong Shin Korea Transport Institute World Bank Transportation Department May 1986 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
ASSESSING THE EFFECTS OF TRUCKING REGULATION IN KOREA
Prepared by: Graham Smith Transportation Department
Tony Michell Hull University, England
Buyong Shin Korea Transport Institute
World Bank Transportation Department May 1986
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
wb350881
Typewritten Text
51346
CURRENCYEQUTVALENT
Currency Unit Won (W)
US$ 1.00 = W800 (end 1983)
US$ 1.25 W1,000
US$ 1 million = W800 million
US$ 1,250 WI million
ABBREVIAnONS AND ACRONYMS
ADT = average daily traffic
KAlST Korea Advanced Institute of Science and Technology
KHC Korea Highway Corporation (toll expressway authority)
KNR = Korea National Railroad
LTL = Less-than-truckload
MOT = Ministry of Transport
OA = Own-account
TK = ton-kIn
IT = trailer-trailer
VOC = vehicle operating cost
Acknowledgement
The authors wish to thank Professor Michael Beesley of the London Business School for his helpful comments and suggestions at each stage in the preparation of this report.
ASSESSING THE EFFECTS OF TRUCKING REGULA nON IN KOREA
Table of Contents
Page No 1. Introduction ............................................................................. 1
2. The Conceptual Framework.......................................................... 2
3. The Korean Trucking Industry ....................................................... 5
4. Trucking Regulations in Korea ....................................................... 9
5. Hypotheses as to Regulations' Effects ............................................. 19
6. Evaluating the Hypotheses ........................................................... 21
7. Conclusions and Recommendations ................................................ 39
Appendix 1: Description of Survey Conducted .......................................... 45
Appendix 2: Details of Cost Calculation ................................................. 49
Appendix 3: Economies of Scale to Fleet Size ........................................... 61
Appendix 4: Information System for Monitoring and Evaluating the Performance of the Trucking Industry ................................. 66
Bibliography ...................................................................................... 73
1. The Truck Fleet by License Type.. . . . . .... . .. . . . . . . .... . . . . . . . .... . .. .. . . .. ..... ... 6
2. Total Road Freight Transport Cost ................................................. 8
3. Key Characteristics of Service Licensing ......................................... 16
4. Size Distribution of Truck Fleets: Korea, Thailand, Brazil and UK Compared ........................................................... 23
5. Seasonal Variation in Truck Traffic, by Province ............................... 31
5
Table of Contents - 2
Al(a) Trucking Output: Fleet Composition and Total Vehicle-Kms .............. ... 41
AI(b) Trucking Output: Load Factors and Ton-Kms Perfo~d ........... ........... 42
Al(c) Trucking Output: Costs per Truck and Fleet Total Costs. ...... ..... ...... ...... 44
A2. Road Freight Consolidation Model, Base Case.................................... 50
A3. Road Freight Consolidation Mode, Terminal Consolidation
(with improved tractor-trailer productivity) ....................................... 58
A4. Operating Costs ofCompany-Owned Trucks in Relation
to Truck Size and Company Fleet Size.. . . . .. . . . . ... .. . . .. .. . . . . .. . .... . . . .. .. . . . . . 64
1. M.O.T. "Area" and "Route" Rates Compared with Truck Operating Costs... 15
2. Truck Direct Operating Costs in Relation to Owner-Operators' Share
of Company Fleet ....................................................................... 38
3. Operating Costs ofCompany-Owned Trucks in Relation to
Company Fleet Size ................................................................. '" 62
ASSESSING THE EFFECTS OF TRUCKING REGULATION IN KOREA
Chapter 1: INTRODUCTION
I. I In 1984 the Korean Ministry of Transport hired consultants from the Korea Advanced Institute for Science and Technology (KAIST) to evaluate the regulations governing its trucking industry. Its objective was to find whether deregulation or changes in regulations could bring down costs, contribute to energy conservation, and serve users better. The study was undertaken in the framework of the World Bank's Highway Sector Loan (2392-KO) in fulfillment ofthe Govemment's Letter ofPolicy Intent.
1.2 The Bank's Transportation Department, which helped supervise the studylL, decided that the Korean experience would be instructive for transportation staff faced with regulatory problems in other borrower countries. This report is the result. It endeavors to summarize the Korean study with attention to the methodology employed and the reasoning behind it, the kinds ofdata collected or desired but missing, and the practical recommendations that emerged.
1.3 The report starts with some key points on objectives and methodology (Chapter 2). It then introduces the Korean trucking industry (Chapter3), with attention to the aggregate costs involved. Chapter 4 sets out the main regulations and their objectives. From this review emerges a series of hypotheses as to the effects of specific regulations (Chapter 5). They are then evaluated in Chapter 6, with particular attention to the marked preference for small trucks observed in Korea and the limited extent of freight consolidation. Finally, conclusions and recommendations for the govemment are drawn in Chapter 7. Appendices deal variously with (a) the surveys conducted for this study, (b) the detailed calculations behind the evaluation of each regulation, and (c) a recommended system for collecting statistics on the trucking industry's performance, as a means to monitor and evaluate regulatory interventions in the future.
KAIST, "Study of Truck Industry Regulation through Survey of Econormc Impact of Transport Regulation on the Trucking Industry", Seoul, 1985.
11
2
Chapter 2: THE CONCEPTUAL FRAMEWORK
2.1 This report tries to demonstrate an approach and consequent line of reasoning that can be followed not only in Korea but elsewhere. In this chapter we explain the objectives and methods which should apply to any study of deregulation in a developing country.
2.2 Studies are meant to lead to action, where shown to be desirable, by governments. The overall government objective which we take as axiomatic is that a change in regulations should be adopted if it will increase the net benefits to be derived from transport activities (accepting, of course, that prescribed safety and environmental standards are adhered to). "Benefits" include not only reduced costs of operation but also service improvements to shippers. These may not always coincide when a given change is implemented. Increased costs, accompanied by improved service levels, can yield net benefits; conversely, a sacrifice in service levels may be justified by a large enough decrease in costs. Users' choice ofmode or type of truck service will be influenced by both the price and quality of services offered.
2.3 Measuring the benefits of better service, in particular, presents considerable practical difficulty, and will often have to be approximated by enlarging the concept of "cost" beyond expenditures on vehicle operation. Thus, shippers' costs may also be considered, e.g. storage and handling, pick-up and delivery and processing through a terminal (where relevant), savings in inventory from increased reliability, decreased loss and damage, etc.
2.4 Transport regulations can directly affect decisions about:
(a) the intensity with which vehicles are utilized, which in tum determines the magnitude of costs allocated to each ton-km performed;
(b) the size of firm and the scope of its operations; (c) the costs of certain inputs, notably wages, overheads and profit; (d) the technology of the vehicle; (e) the services offered.
2.5 To determine whether a change in regulations will be beneficial or not, one has first to establish a cause-and-effect link between the proposed change and relevant decisions by shippers and operators. Such a link cannot be taken for granted. Regulations may exist on paper but not be enforced. Efforts at enforcement may be too weak to affect behavior. Or other constraints on behavior may be masking potential effects of transport regulations --for example, bans on the import ofvehicles. In the terminology of mathematical programming, the regulation may not be a binding constraint.
2.6 Assessing the effect of a regulation on a given category of carriers will normally involve comparing them with other categories in the same country or with similar carriers in other countries who are not regulated. The cause-and-effect question requires that:
3
(a) differences in utilization rates, scale and scope of operations, factor costs or technology among groups of truckers subject to different regulatory regimes have to be established;
(b) possible explanations other than regulation have to be identified, and
(c) the effects of these other factors have to be assessed independently, so that residual differences can be attributed to the difference in regulation.
2.7 If standardization for the other causal factors explains the entire observed difference between the more regulated condition and the less, one has to conclude that the regulation has no effect, i.e., is not binding.
2.8 What is an acceptable standard ofproof? There is no clear standard, but the arguments for change have to be persuasive enough to overcome the objections of those whom the existing regulations protect.
2.9 In estimating the value of potential benefits there are useful parallels between this exercise and the cost-benefit analysis of roads, with which many readers will be familiar. There is also one important difference.
2.10 Reductions in vehicle operating costs per ton are beneficial to existing traffic using that mode. (We use 'mode' here to cover the various license categories of for-hire trucks and own-account trucks.) This is the counterpart ofVOC savings to existing traffic on a road being improved. Savings in the resource cost of trucking can readily be achieved by raising truck productivity so that standing costs are spread over more ton-km performed. If competitive pressures push freight rates down, existing traffic pays less.
2.11 If rates for one commercial truck category go down, some shippers now using other 'modes' (including own-account haulage) will switch and make a saving. A reduction in the costs ofown-account carriage will likewise encourage mode-switching. This is the counterpart of diverted traffic in road cost-benefit analysis. The saving is the difference between the generalized cost by the original mode and the 'with-change' generalized cost by the 'new' mode. It is less than the saving to existing users of the 'new' mode; if the cost by the 'old' mode is not known, a rough estimate of the saving per diverted ton oftraffic would be half the saving per 'existing' ton. As for estimating the quantity that would divert, disaggregated analysis of specific haul distances and consignment sizes will be needed.
2.12 The benefits to existing and diverted traffic as described above assume a fixed trip matrix. Additional (though usually smaller) benefits will accrue to traffic generated by the lowering ofrates. This is the counterpart of generated traffic in road CBA. The quantity of additional traffic depends on the price elasticity of freight transport demand (which for most commodities is low). The saving per ton will follow the rule of half. 2.13 The major difference with road CBA is the possibility that there are monopoly rents, or inflated margins between the resource cost to the trucker and the rate he charges the shipper. These may take the form of excess profits or they may be soaked up in inefficient production methods, or higher quality services. If the dropping of restrictions on entry, for example, causes these margins to be squeezed, the immediate saving to the shipper is in the nature of a transfer: the shipper's gain is the trucker's loss. It is not per se
4
a net benefit to the economy at large. On the other hand it will generate additional traffic and probably also trigger management action to lower costs, which can be a substantial benefit.
2.14 In terms of investigative effort, it pays to concentrate on selected parts of the whole regulatory system where benefits promise to be greatest. These will be where:
(a) a large part of the whole freight transport market is at stake;
(b) the potential benefit per ton or ton-km appears large; and
(c) the cause and effect linkage appears strong.
2.15 The different determinants of transport cost --productivity, scale and scope of output, factor prices and technology-- are likely to reward study in differing degrees. Productivity and utilization improvements are essentially pure gain; truck capacity is mostly an unstorable commodity --if it is not sold it is lost. Moreover, intensity of use is relatively easy to document. Cuts in factor prices, in contrast, are not gain to both sellers and buyers; the increase in welfare is the producers' or consumers' surplus derived from generated traffic (the Marshallian triangle). Nor are prospective cuts in factor prices nearly so easy to assess. For one thing most beneficiaries ofmonopoly rents take pains to hide them. For another, evaluating the welfare increase requires knowledge of both supply and demand elasticities, which are elusive. Comparative evaluation of different price/service quality combinations is very difficult. Scale and scope ofoperations (company size or network extent) are easy to document but reliable cost information is hard to obtain; isolating the effects of regulations from other determinants of scale and scope also demands care. The impact on technology is even harder to evaluate: first to isolate the effect of regulations from other causal factors and then to predict response to a regulatory change. Furthermore, technological improvements such as special-purpose truck bodies may raise rather than lower the more readily measured elements of transport cost, but be justified by savings in the harder-to-evaluate aspects like reliability and safety.
2. 16 Finally, to conclude that it is very hard to nail down the effects of regulations does not mean that they are insignificant or not worth pursing. Total expenditures on freight transport by road are very large --in Korea about $4.2 billion per year, equivalent to 5% of GNP. Every one percent reduction would be worth $42 million each year. The evidence suggests that savings of several percentage points are feasible, through changes in regulations whose costs would be a small fraction ofthe benefits.
5
Chapter 3: THE KOREAN TRUCKING INDUSTRY
How trucking is organized in Korea
3.1 In Korea, as in most countries, trucking is divided into own-account and for-hire, referred to in Korean as private and commercial. Each is licensed to do only the one or the other. Commercial trucks are further divided by licensing into four categories: area, route, special and "yongdal" (taxi-style pick-ups).
3.2 Area trucks can ply for hire within a province, and can carry to other provinces or from another province back to its home province, but not from a second province to a third. They are not restricted as to commodity. They are registered either as general freight vehicles (mostly flat beds and open-tops) or as dumpers. Special trucks are those with specialized bodies: articulated (tractor and semi-trailer), tankers and refrigerated vans (whatever their size), and miscellaneous others. They can ply throughout the country. Route trucks operate "freight trains on roads": a scheduled frequency service on specified routes. In practice in Korea they are essentially a parcels service.
3.3 Yongdals are pickups with a taxi meter, licensed to carry small loads (up to one ton). The owner of the shipment may ride with it if he wishes. Yongdals can operate within a city or province, subject to the same geographical restriction as area trucks.
3.4 Operating licenses further distinguish two classes of commercial owners: companies and owner-operators. From the early 1960's until the 1980's owner-operators were not allowed to operate independently, but were required to affiliate with a company. This controversial requirement is taken up in the next chapter.
Fleet growth
3.5 Private trucks have recently grown in number far faster than commercial trucks: 21 % per year from 1979 to 1983 compared to 6.5%. The growth ofboth categories has been most rapid among vehicles under 2 ton capacity, but the differential in growth rates has held throughout the size spectnun.
3.6 The number of trucks and companies by license type is as follows:
6
TABLE 1: THE TRUCKF'LEETBY LICENSE TYPE (END 1983)
Commercial
No. of companies No %
No. of trucks No (000) %
%of Grand
Avg. Trucks per Company
Area Route Special
of which Tractor-trailers
864 25
344
(44)
70 2
28
(4)
38.7 1.3 4.1
(3.5ai)
88 3 9
(8)
13 0 1
(1)
44 53 30
80
Total T,Z33""1 44.T lUlJ 14 36
Y ongdals « 1 ton capacity) of which companies 771
owner-operators 14,237 Total commercial
17.6 (3.5)
lliJ..) 61.7
6 (1)
..ill 20
Own-Account
> 1 ton capacity <1 ton capacity Totalown-aecount
115.7ai 114.9 230.6
38 37 75
Government & miscellaneous 13.6 4
Grand Total Fleet 305.8 Tau-
ill Excludes trailers (5,500 commercial, 2,200 private) bl Nearly all special licenses are held by area companies, and most route companies also hold area licenses. Furthermore, a company may hold licenses in more than one area. The total number ofcompanies is therefore no more than about 800.
7
Market segmentation
3.7 The freight transport market in Korea, as in most countries, is highly segmented. The five key characteristics of demand that account for this segmentation are:
(a) consignment (b) time sensitivity of the freight (c) length of haul (d) requirements for special-purpose truck bodies (e) weekly or monthly volume to be shipped.
3.8 Haul distances are relatively short. The average for inter-urban trucking is only about 100 km, and Seoul to Pusan, one of the longest hauls, is under 450 km. Rail is a serious competitor for hauls beyond 150 km, and has a 40-50% share of such traffic.
3.09 In all market segments shippers have a choice between own-account trucks and commercial area trucks. For large consignments, there is the third option of "specials" (tractor-trailers), while for small consignments the further options are Yongdals for short distances and "route" firms for longer distances. Route companies are the only commercial carriers to provide a consolidation service. If a shipper is willing or able to consolidate loads himself, he may have a choice among all five categories.
Truck Sizes
3.10 The truck fleet, very largely of domestic manufacture, heavily favors four sizes: 2.5 tons capacity, 5.5 tons, 8 tons, and 11 tonsY. Tractor-trailers are conspicuously few in number: among commercial trucks they are only 8% and including private tractor-trailers they are less than 3% of the total fleet (exceeding one ton). They are concentrated in three market segments: commercial carriage of containers in import/export trade, commercial carriage of sheet metal, wire and girders, etc., on flatbeds, and private carriage of bulk cement. There are no box bodies or vans larger than 4.5 tons. No containers are used in domestic movements, nor have swap bodies been introduced. On the other hand, full-trailer combinations of an II-ton truck hauling a 10-ton trailer have recently been introduced and are growing quickly in number. The surprising absence of semi -trailer types found in abundance in other countries at a similar stage ofdevelopment is one ofthe major concerns of this study.
Total User Cost of Freight Transport by Road
3.11 A rough estimate of the total freight transport cost to users is essential, to establish a sense ofproportion among different market segments and to suggest the size ofpotential benefits. The total cost is about $4.2 billion (Table 2), divided into two-thirds own-account, one third commercial.
The corresponding gross weights are about 5.5 tons, 9 tons, 15 tons, and 19 tons.
8
TABLE 2: TOTAL ROAD FREIGHT TRANSPORT COST
Billion won Million US$ % of total a)
Suppliers' costs Own-account 2,220 2,770 66 Commercial 1,030 1,240 30
Subtotal 3,250 4,060 96
Commercial carriers' profit (3%) 30 40 1
Brokers' commissions 60 70 2 (5.7%) 60 70 2 Value added tax (6%)
3,400 4,240 100 Total users' costs
(GNP, 1983) (80,310)
a)US$ 800 (end 1983)
The share of the various license categories in the total trucking market is as follows:
%of %of Ton-km Supply Cost
Own-account 46 68 Area 40 22 Route 2 1 Special 11 4 Yongdal 1 5
Total 100 100
A detailed breakdown is given in Appendix Table AI.
9
Chapter 4: TRUCKING REGULATIONS IN KOREA
Motives for regulation
4.1 Korean regulation of trucking reflects several distinct motives. The principal are as follows:
(a) Controlling the industry to permit rapid mobilization in the event ofan emergency. This is a pervasive and dominant concern throughout the South Korean economy, because of the continuously hostile and tense relations with North Korea.
(b) Stabilizing the trucking market against cyclical swings. Transport regulators in many countries appear to believe that all truckers are either crooks or fools; the customers need to be protected from the crooks and the fools need to be protected from themselves. Something of this attitude can be seen in Korea, also. The MOT's own interpretation is that it has an obligation to protect truckers from bankruptcy by awarding no more new licenses than can be justified by its expectation of future demand growth. Conversely, it should protect shippers' interests by encouraging moderate growth of truck supply by setting targets.
(c) Keeping prices down In an economy where government control of pricing is widespread, Korean truck regulators aim to protect shippers from abuse of monopoly power by truckers, as well as to control inflation, by setting maximum rates. This concern was more potent in the early days of Korean trucking, i.e. up to the mid-1970s. Notwithstanding its concern for market stability and protection of the railway, it has not enforced these rates as minima.
(d) Restricting imports. In the 1970's Korea suffered from a chronic balance of payments deficit, while also wishing to build up its own nascent motor vehicle industry. Consequently imports of vehicles have, until the mid-1980's, been subject to severe restrictions (special import licenses).
(e) Promoting economies of scale. The concept "big is beautiful" has many supporters in Korea. MOT has firmly believed that the larger a transport enterprise, whether taxi, bus or truck company, the more efficient it should be.
(f) Fostering balanced regional development: In framing legislation MOT has always been concerned that backward regions (i.e. the southwest and northeast, away from the Seoul-Pusan corridor) should be well served, and has imposed routes and frequencies on "route" truck companies to this end.
(g) Protecting the railway from road competition. Korean National Railways (KNR) had a virtual monopoly on long-distance traffic until the 1970's, when paved roads started to be built on a large scale. Trucking's rapid expansion since then has been seen as a threat to the government's investments in KNR
10
(h) Rewarding officials for services to the State: As in several countries, the government in Korea has used truck licenses as rewards to public officials and military leaders upon their retirement.
(i) Protecting pavements: The Ministry of Construction and the Korea Highway Corporation (the government-owned toll expressway authority) are concerned to control vehicle weights to protect pavements.
(j) Controlling urban congestion: The population of Seoul is now approaching 10 million, growing at 3-4% per year. Pusan has 3.5 million inhabitants. Congestion is a major concern.
4.2 Absent from this list of motives underlying Korean trucking regulations is road safety, which MOT officials do not yet regard as a high priority.
4.3 How well the regulations achieve the above objectives is wide open to debate. The analysis in Chapter 6 will endeavor to suggest some answers.
Principal legislation
4.4 Modem regulation of trucking in Korea dates from 1961, when the Vehicle Transport Business Law was passed. It defined the four categories into which all commercial trucking is required to fall and empowered MOT to set maximum freight rates. Legislation is interpreted by the Government in a traditional Korean confucian tradition, in keeping with which a government official can interpret the law according to his educated opinion without regard to precedent or to other officials. Officials at national and provincial level have fairly extensive discretionary powers. This means that it is often difficult to pinpoint the legislative origin of certain practices and attitudes that have become widespread.
Regulations common to all truck operators
4.5 Business license: Anyone wishing to operate a truck needs a business license, either as a transport firm or as a manufacturer, wholesaler, farmer, etc., carrying for his own account.
4.6 Maximum age of truck All trucks must be scrapped after 13 years.
4.7 Containers: Containers may be used only in import/export business, not in domestic trade. The first containers used in Korea were themselves imported, and were permitted to enter the country under customs bond for re-export only. This practice persists even though containers are now manufactured in Korea.
4.8 Restrictions on movements: Since 1979 all trucks over 2.5 tons have been banned from circulating within central Seoul throughout working hours, to help relieve congestion. There are complex rules allowing some access on designated routes, but the general effect is to push truck arrivals and departures into the night.
II
4.9 Another restriction on where trucks may operate concerns pick-up and delivery to and from railway freight yards. KNR used to operate such services itself exclusively, but in 1968 they were sold off to form Korea Express, now the largest commercial carrier. Since then some 35 other companies have been licensed to deliver specific commodities to and from specific stations.
Regulations applying to all operators but differentially on own-account and commercial
4.10 Imports: Imports of trucks are subject not only to customs duties but also to quantity restrictions. Special import permits have been granted in moderate numbers to commercial trucking companies (on the basis of a case-by-case evaluation) but not to own-account operators. While this was very restrictive in the 1970's, when only a limited range ofmodels was assembled domestically, it has become progressively less so in the 1980's as domestic manufacturers have expanded the quantity and diversity of their output. The quantity restrictions are to be abolished in July 1986.
4.11 Taxation: Besides a number of taxes common to own-account and commercial trucks, commercials are assessed VAT at 6%, not on actual revenues but on a govemment estimate ofwhat the volume ofbusiness ought to be.
4.12 Other significant differences in the treatment of commercial and own-account operators are as follows:
Commercial
(a) Insurance Compulsory liability: Compulsory liability: 37,100 62,650 won won Comprehensive: Comprehensive: optional 552,500 won
(b) Penalty for 10 days suspension of license 10 days suspension of license + overloading + 120,000 won fine 20,000 won fine
(c) Vehicle inspection Twice a year Once a year
Regulations applying only to commercial operators
4.13 Entry into commercial trucking is subject to quantitative restrictions, whereas own-account trucking is not. An own-account operator, on purchasing a truck, needs only to register it, which authorizes him to use it for carrying goods needed in his business, but no others. (A farmer, for instance, is not allowed to transport his furniture ifhe moves house.)
4.14 Entry quotas: A would-be commercial trucker needs two kinds of license: one to operate an approved type of service, another for each new vehicle he wishes to purchase, specified as to size. The applications must satisfY MOT's supply criteria, but it has never
12
legally defined them, successive officials using their judgement on an ad hoc basis, partly depending on the lobbying power of the existing operators. In the early 1980's MOT considered the need proven if the average distance driven by existing trucks rose above 170 kms per day. Since this infonnation had to be obtained from established operators, it became difficult for new companies to enter the market.
4.15 In 1983 the power to decide the number ofvehicles in "area" fleets was delegated to provincial and "special city" governments. These local governments have applied very different criteria, some granting all requests, others applying a fonnula based on GNP growth. The general practice appears to be that the authorities once a year announce their intended quota for the coming year and invite applications up to a deadline date. An applicant who is refused or misses the deadline then has to wait a year until the next opportunity. In practice new applications have been awarded in fits and starts not closely correlated with the business cycle.
4.16 The same company may hold licenses for more than one kind of service (e.g., area and special), and may hold area licenses in more than one province. This does not mean, however, that the same truck can serve all areas for which the company is licensed; the vehicle license is specific to one province or special city. "Area", "special" and "route" company licenses are not transferable, while the vehicle licenses, including those for yongdals, are transferable, but only to other owners affiliated with the same company, or to the company itself.
4.17 Owner-operators: The 1961 law created the "ji-ip" license, requiring all previously independent owner-operators to associate with trucking companies oftheir choice so that all fleets should have a prescribed minimum number of trucks. This fonn of ownership is not exclusive to anyone license category but is found among area, route and special fleets. The minimum fleet-size criteria to be met at the time of entry are as follows:
Seoul, Pusan Other Cities Province
Route truck 30 30 30 Area truck <8T 20 10 10 Area truck >8T 20 5 5 Special truck No minimum limit Yongdal No minimum limit
An application can also by refused if the operator fails to provide minimum parking or garage space.
4.18 Each provincial government, in consultation with the local truckers association, sets the fees that owner-operators must pay to the company to which they are affiliated. They are revised every few years. The fees are higher for larger trucks, but are unifonn for all trucks of a given size within the province or city. The fees are highest in Seoul and lower in the less urbanized provinces.
4.19 In 1980 owner-operators ofyongdals were allowed to operate independently for the first time, whereas in 1981 orders were passed to integrate all larger ji-ip trucks fully
13
into the company fleets by requiring each company to buyout a prescribed number of their owner-operators each year. The latter move was resisted by both company owners and owner-operators and in August 1985 the MOT rescinded the order. Owner-operators of trucks under 5 tons are now permitted to operate independently. However, a four-year transition was stipulated for existing license-holders; one quarter of those applying will be "set free" each year. This is to give the companies time to adjust, by buying an off- setting number of trucks of their own, if they so wish.
4.20 Operating area: In the 1960's the area for which "area" trucks were licensed was interpreted as a single county, a subdivision of a province. During the 1970's, the interpretation of"area" was progressively expanded to include a whole province. In the case of the four "special cities" (Seoul, Pusan, Inchon and Taegu) their area now extends to the province in which they are located.
4.21 Route changes: Each route company has a route network authorized by the MOT. Both entry to new routes and withdrawal from existing ones has been formally blocked for several years, as the MOT has refused to approve any changes, to protect the principle that unprofitable routes should be cross-subsidized from profitable ones. In practice firms have resorted to some unauthorized route changes; there has been no enforcement.
4.22 Driver education: Commercial drivers are required to receive training within their company once a month and to undergo practical training twice a year.
Rate Regulation
4.23 The MOT issues one tariff for "area" carriers and traffic other than containers on "special" trucks, another for "route" traffic, and a third for containers. The area tariff specifies the maximum rate by truck capacity and distance, permits surcharges for particular types of commodity and allows discounts for contracts covering long hauls and lasting over 3 months. While the "area"I"special" rates taper with distance, appropriately reflecting the structure of operating costs, "route" rates do not. They merely extend a uniform rate of 93 won (about 11 ¢) per ton-km to any shipment size, any distance or any route. The rates have not been revised since 1981, despite inflation of about 40% from 1981 through 1985, and were clearly set high for all but "route" traffic.
4.24 In practice "area" rates are set by market forces, with carriers commonly negotiating one-year contracts at 60-80% of the MOT rate. The Government does not enforce the maxima in any way. Route and container rates are not normally discounted, though some discounting of 10-20% was reported in both segments in early 1986, attributed to competitive pressures in a slack market. The area tariff presumes that the shipper pays a full truckload rate regardless ofthe shipment size. However, some contracts are known to be on a per-ton-Ioaded basis.
4.25 The route tariff bears no relationship to the structure of costs. A lO-kg parcel sent from one rural village to another 10 km away will be charged the same per ton-km as a two-ton shipment going from Seoul to Pusan, even though the scale economies of handling, distance and traffic density will make the smal1fshortlrural movement many times more costly per ton-km. The only scope for route companies to adjust is to use smaller trucks on
14
the rural routes (4.5-tonners instead of the normal 8-tonners), and to discount only on larger shipments. The "route" rate is lower than the "area" rate for consignments of less than two tons over any distance. For three tons the break-even distance is about 100 kIn, and for 10 tons, it is 25-30 kIn (Figure 1).
4.26 On average every year one company with a "route" license has ceased its operations, lending credence to their complaint that the "route" rates are too low, particularly on the low-traffic routes they are obliged to serve. This includes one of the most efficient and cost-conscious firms, Korean Express. It also implies that there are enforcement mechanisms, ifnothing other than public outcry, that deter them from violating the maxima.
4.27 The container rates are well above those for non-container tractor-trailer movements. Since containers are used only for import/export traffic, the effect of the high rate is to encourage shippers to stuff and strip containers at the port and move goods inland by an 8-ton or II-ton "area" truck. It is anomalous that tractor-trailers carrying containers should charge more than 8- or II-ton "area" trucks, while their greater size should in principle allow them to operate at lower cost per ton-kIn performed. In practice their average costs are not much lower, because their load factors are lower and they fail to achieve higher annual mileage, and their rates are not attractive. The absence of common-user consolidation terminals is thought to contribute to the lack-luster performance of tractor-trailers. More will be said on this in Chapter.6.
4.28 The key differences among the different service license types are summarized in Table 3.
Brokers
4.29 Brokers perform a number of important functions: matching up truckers' supply and shippers' demands, intermediation in carriage contracts, billing and the provision of working capital to truckers. Freight brokers in Korea can act both as agents for contracts between shipper and trucker, and as a contracting carrier vis-a-vis the shipper, separately subcontracting with a trucker to perform the haul. As agent they charge a commission; as contracting carrier their income is the difference between the transport rate they negotiate with the shipper and the rate they negotiate with the trucker. There is no legal requirement for a waybill.
15
Figure 1
M.O.T. "Area" and "Route" Rates Compared with Truck Operating Costs
[Not available electronically]
16
TABLE 3: KEy CHARACTERISTICS OF SERVICE LICENSING
Service specifications (a) Nature of service
(b) Contract applies to:
2 Vehicle specifications (a) Nature
(b) Size (payload capacity)
(c) Required equipment
3 Geographical constraints
4 Need criteria (annual quota)
5 Rate regulation
Area
Ad hoc & term contract
Whole vehicle
Non-specialized or dumper
>1 ton
None
Either origin or destination must be in home proVInce
Yes
Maxima
Route
Scheduled
Consignment
Nonspecialized
>1 ton
None
Licensed routes only
Yes
Obligatory
Special Yongdal
Ad hoc & Term Contract
Ad hoc
Whole Vehicle
Whole vehicle
Tractor-trailer, tanker, or refrigerated
Nonspecialized
>1 ton <1 ton
None Taxi meter
Unrestricted Either origin or destination must be in home proVInce
Yes Yes
Maxima Obligatory
4.30 Very few owner-operators rely on the company to which they are affiliated to find them loads; most find their own. To the latter the brokers are an important source of business. The truckers, however, bitterly accuse the brokers ofexcessive margins and dishonest practices. In response the Government has tightened up on certain controls since 1981. To qualify for a license to do business, brokers now have to satisfy requirements as to a minimum size of office and capital. However, these are not hard to satisfy: 33rrf space and $35,000 equity. An exception is the city ofTaegu, where until 1984 lobbying by truck companies prevented any brokers' licenses being issued. Brokers operated there illegally instead. The maximum fee permitted by regulation is 5%, but fees up to 20% of the total rate are commonly reported, and the truckers' complaints have not abated.
17
4.31 By 1984, 995 brokers' licenses had been issued, almost double the number three years earlier. Many of them are trucking companies ("area" and "special"); the brokerage activity is captive to the trucking firm and is no different from an integrated marketing and accounts department. Others are one-man, one-desk, one-telephone operations serving primarily out-of-town truckers, who by definition are mostly ji-ip operators. This is particularly the case in Seoul. For them the brokers make a market in backhauls at rates below the normal front-haul rates.
4.32 When the brokers act not as agents but as carriers, they can in uncompetitive markets cut the rate negotiated with the trucker down to the trucker's (near-zero) marginal cost, while negotiating a full rate with the shipper. But in practice competition among brokers places limits on this. There are definitely too many small brokers to speak of monopoly power, nor is there any evidence that they collude as a cartel. One suspects that the truckers' accusations arise more from the competitive pressure the brokers exert to keep front-haul rates down, and to the vulnerability of out-of-province "area" truckers, who have to choose between an illegal backhaul to a third jurisdiction or a legal ride home empty.
Enforcement agencies
4.33 The regulatory agencies are the Freight Division ofthe MOT's Land Transport Bureau (specifically two or three individuals) and the Transport Division of the Commerce and Industry Bureau in each province or city (typically one or two individuals).
4.34 The trucking profession also assists the authorities in monitoring the activities of its own members. All commercial truckers belong --at least nominally-- to their provincial truck association, a private body recognized by the government for the paramilitary mobilization of the trucking industry. The associations also lobby on behalf of their members.
4.35 The following sets out who enforces what:
Business license: MOT through provincial governments and truck associations
Vehicle license: MOT through the provincial governments
Illegal commercial acts: Police, truck associations and (e.g. a private truck provincial officials carrying for hire)
Overloading: MOC through regional offices and KHC at tollplaza weighbridges
Vehicle safety: Police
Penalties and enforcement
18
4.36 Different govemment officials have the power to levy certain fines or penalties ranging from suspending the license for a specified vehicle for 10 days, upto complete revocation of the license to do business. This is independent ofjudicial process. The police have similar powers. This frees both sets of officials from having to prove their case in a court of law.
4.37 There were almost 7,000 cases in 1983 where private trucks were accused of operating commercially, or 3% of the total private fleet. Two thirds resulted in suspension of operations for up to six months and a fme. By 1985 the number of offenders had not changed substantially, though fewer than 40% were penalized. The National Truck Association also reports about 600 overloading violations in 1985.
19
Chapter 5: HYPOTHESES AS TO REGULATIONS' EFFECTS
5.1 The main cost and quality parameters that regulations can impinge on are:
Productivity indicators
(1) Annual kms driven (2) Empty running (3) Load factor on loaded trips
Scale and scope of operation
(4) Truck size (5) Fleet size (6) Number of consolidation terminals (7) Scale of each terminal
Factor costs
(8) Overheads and profits (9) Wages
Vehicle technology
(10) Truck design (plain/special-purpose) (11) Fuel efficiency
Service characteristics
(12) Speed: freedom to charge premium for faster service (13) Reliability: requiring informed contracting between trucker and shipper
5.2 From the foregoing assessment of the current Korean situation, the following hypotheses emerged, grouped according to the nature of the presumed effect.
Productivity effects
(1) The ban on part-loading by "area" trucks, entry restrictions on "special" trucks, the fact that "area" licenses are inflexible as to truck size, and the ban on the use of containers in domestic trade together discourage use of tractor-trailers, and distort the truck size distribution to smaller trucks, thereby raising costs and inhibiting service innovation.
20
(2) "Area" restrictions on back-hauls prevent trucks from migrating to provinces experiencing seasonal peaks of demand, thereby a creating excess capacity and raising costs.
(3) The ban on own-account trucks operating for hire raises costs by preventing use of empty capacity on backhauls.
Scale and scope effects
(4) The minimum fleet size requirement andji-ip system protect high-cost company-managed fleets (raise overheads and profit or allow inefficient operations to remain profitable).
Factor cost effects
(5) The entry/quantity restrictions on yongdals likewise allow operators to raise profit margins (create oligopoly rents).
Each of these hypotheses is evaluated in the next chapter.
5.3 A further hypothesis reflecting views of the protagonists of regulation is that larger fleets offer economies of scale. Its evaluation is included as Appendix 3.
21
Chapter 6: EVALUATING THE HYPOTHESES
6.1 In this chapter we attempt to assess the evidence that supports or refutes the hypotheses set out in the previous chapter and to evaluate the economic cost of the various regulations. It will be seen that the evidence is ambiguous or unreliable in most cases. It is therefore wrong to suggest any certainty in the cost of a given regulation's effect. What can instead be said is that where, based on reasonable assumptions, the cost appears large, closer examination is warranted.
6.2 Details of the cost calculations can be found in Appendix 2. In addition to the KAIST regulation study, these estimates draw extensively on a subsequent study offreight terminals in Korea, also carried out by KAIST under the same World Bank project'/. It needs to be stressed that the following estimates assume the regulations are fully adhered to. Where the effect is on the vehicle stock, this is reasonable; there is little doubt as to whether vehicles do or do not exist. Where operating restrictions are concerned, on the other hand, conclusions must be cautious: the more a regulation is disregarded in practice, the more the estimate will over-state the economic gain to be had from abolishing it.
A. Ban on Part-Loading by Area and Special Trucks, Entry Restrictions on Tractor-Trailars, and Ban on Use of Containers in Domestic Trade
Background
6.3 Several factors, both regulatory and other, have the combined effect ofdeterring shippers from consolidating loads in large trucks. Disentangling the effects of each factor is infeasible. We have to consider the benefits that might result from removing three restrictions at more or less the same time.
6.4 One of these is the ban on part-loading. Owners of "area" and "special" trucks charge shippers for a full truckload rather than the tonnage or volume carried. While part loading is not explicitly forbidden, it is understood to be precluded by interpretation of the basic Vehicle Transport Business Law, which reserves less-than-truckload business for "route" companies. Even though this ban is hard to enforce, it appears to be observed widely; hiring by the exclusive truckload has become standard commercial practice.
6.5 A second restriction is that purchase of a tractor-trailer for commercial use requires a "special" license. The licensing authority evaluates applications once a year against its perception of need. Holders of existing "area" licenses on smaller trucks are not allowed to trade up to a tractor-trailer other than by obtaining a "special" license. Route Services may not use TT's (the fine for doing so is 240,000 won (US$ 300) per violation). 6.6 Thirdly, containers may be used only in export-import trade, not for domestic transport.
31 KAIST/ASG Transport Development AB, "Study on Improvement of the Freight Transport System," Interim Report, Seoul, January 1986.
22
Hypothesis
6.7 The combined effect of these three restrictions is to distort the size distribution of the fleet towards small trucks, so that the commercial trucking industry cannot take advantage of the economies of scale of larger vehicles or the technology of detachable trailers and load consolidation.
Evidence
6.8 The most cogent evidence of regulation restricting ownership oflarge trucks would be: (a) above-nonnal profits for the few who do own them, (b) premia on the resale value of "special" licenses, or (c) a high refusal rate among applicants for "special" licenses. In fact evidence of this kind is weak or absent.
6.9 As regards above-nonnal profits, no direct data are available. Indirect evidence is that the MOT rates for TT's (both the special tariff for containers and the regular tariff for other business) exceed our synthesized estimates of operators' costs by a much wider margin than the rates for smaller trucks. While the official rates are irrelevant for much repetitive business carried on contract at rates discounted from the official tariff, they do discourage shippers with occasional business from using TT's. We have been unable to establish whether "specials" discount less than "area" trucks.
6.10 Company licenses for "specials" are not transferable, so there are no legally admissible resale premia. As for license refusals, no statistics are published. Officials in selected provinces, when interviewed in 1985 and 1986, said that all "special" applications had been granted. At the same time it seems probable that in the Korean cultural context overt refusals would as far as possible be avoided. Infonnal negotiations before the annual deadline would discourage a prospective applicant from applying fonnally if he knew he would be refused. It is also thought that quotas for "specials" have been more liberal since 1984, when licensing was decentralized from the MOT to the provinces, but in the absence of statistics on refusals, this evidence is only anecdotal. Finally, in 1986, since owner-operators of trucks smaller than 5 tons capacity have been "set free", the companies to which they were fonnerly affiliated have shown a preference for larger trucks when replacing those "lost"; one third have been larger than 5 tons.
6.11 The only strong evidence comes from other countries. Comparison with Thailand, Brazil and the United Kingdom, essentially unregulated countries as far as trucking is concerned, suggests that the popularity of light own-account trucks in Korea is indeed unusual (Table 4). Such trucks, mostly with a 2.5-ton capacity (about 5 tons gross weight) make up over half of all trucks in Korea, whereas in the UK they are a little over a quarter of the entire fleet, in Thailand less than a quarter, and in Brazil only a sixth. (The truck fleet is defined to exclude pick-ups, small vans and any other vehicles with a payload capacity of less than one ton.) The comparison also shows that it is among
TABLE 4: SIZE DISTRIBUTION OF TRUCK FLEETS: KOREA, THAILAND, BRAZIL AND UK COMPARED
23
United Korea Thailand Brazil Kingdom
Per capita income (1983) 2,010 810 1,890 9,050 Population density (per km2
) 408 96 15 228 Average road haul distance (km) 117 305 365 73
Own-account
Light 72 27 30 32 Medium 18 53 44 38 Heavy 9 19 22 8 Tractor-trailer 1 4 12
Total 100 100 100 100
Commercial
Light 26 28 14 16 Medium 17 37 33 33 Heavy 49 36 41 13 Tractor-trailer 8 12 38
Total 100 100 100 100
Own Account as % of total fleet: 72 82 56 66
-_..
Definitions:
Korea Thailand Light 1-3.0 ton cap. 4 wheels Medium 3.1-7.0 6 wheels Heavy >7 ton rigid 10 wheels Very Heavy
artic. N/A
Brazil 1-4.0 ton cap.
4.1-8.0 8.1-15.0
>15.0
u.K. <7.5 tonG7.5-17 >17 tons
artic.
VW " "
Sources: Korea: Present study Thailand: Kampsax International, "Study ofTrucking Industry," Phase I
Final Report, Bangkok, 1983 (for Ministry of Communications, Thailand)
Brazil: Empresa Brasileira de Planejamento de Transportes (GEIPOT), "Estudo sobre 0 Transporte Rodoviario de Carga," Final Report, Brasilia, 1984 (for Federal Ministry of Transport, Brazil).
UK: Department of Transport, "The Transport of Goods by Road in Great Britain, 1984", London, 1985
24
25
own-account trucks that the preference for small trucks in Korea is most marked; in the commercial fleet small trucks make up virtually the same share in Korea as in Thailand.
6.12 A point worth noting from this and other international comparisons is that smaller trucks are correlated with shorter hauls. One explanation is that achieving a full load on a larger truck in some circumstances entails additional costs that are invariate with haul distance. In such circumstances the larger truck, whose cost per ton-krn is lower, is cheaper overall only on hauls beyond a break-even distance. Thus it comes as no surprise that small trucks should be least numerous, relatively speaking, in Brazil, where the average haul distance is longest. It is for this reason that the UK is included as a comparator, since its average haul distance is shorter than Korea's. At the same time, clearly, its income level is much higher, and the higher cost of labor may lead operators to prefer larger (and hence fewer) trucks as a way to save on drivers' wages. In short, some part of Korea's preference for small trucks can no doubt be attributed to the short average haul, but how much is hard to say.
6.13 Why then, do Korean owners and customers prefer small trucks? The immediate answer seems to be the ban on part-loading combined with a preference for small consignments. Consignment size is largely determined by the buyer of the goods, on the basis of several considerations, one of which is transport cost. If goods were dispatched less frequently, consignments could be larger. The question could therefore be recast as: What discourages larger consignments, either by shippers sending out less frequently or by operators consolidating loads from multiple shippers? The following factors probably all contribute:
Demand
(a) Much of Korean business is small-scale, so that total volumes to be shipped from each source per week or month are not large.
(b) Distances between producers and consumers, or the next tier in the distribution pyramid, are short, so that transport costs are not a large part of the product price; other distribution characteristics weigh more heavily (reliability, sales back-up, etc.).
(c) Specifically, shippers prefer to avoid storage, or their customers do, so they dispatch small loads frequently. The terminal study cites storage costs and inventory carrying costs in the range 800-7,000 won per ton per day (US$ 19). The lion's share of this is inventory carrying cost. Short-term working capital costs 2-3% per month (25-35% per year). At this level it appears rational to send out high-value goods, of the order of $5,000 per ton and above, frequently in small shipments, but other explanations need to be sought as to lower-value commodities. Ignorance of the relevant costs may play some part.
26
(a) Large trucks have problems of access in many locations, both to small commercial establishments in towns and to rural areas where the terrain is rugged and many roads are still unpaved.
(b) Only route companies are allowed, by regulation, to consolidate loads from multiple shippers.
(c) The structure of "route" rates (para 4.25) and the license obligation to operate scheduled services even on low-traffic routes force route companies to crosssubsidize from large shipments/long hauls/high-density routes to small shipments/short hauls/low-density routes. They therefore over-stimulate the latter by under-pricing and discourage the former by over-pricing.
(d) Compounding this distortion, route companies are not allowed to operate tractor-trailers, since the latter require a "special" license. "Route" and "special" licenses are mutually exc1usive4/
• Route companies therefore cannot lower their costs on high-density routes as they otherwise might. (Full trailers are catching on as a second-best substitute.)
(e) Mid-sized shipments (2-8 tons) therefore go to private trucks or commercial "area" trucks.
(f) Private truckers are not allowed to consolidate other people's loads, so prefer to own small trucks.
(g) Commercial "area" trucks likewise are not allowed to consolidate loads from multiple shippers, either through terminals or by multiple pick-ups in a doorto-door trip. Shippers hire the smallest truck into which their load will fit.
(h) "Special" trucks, which are mostly tractor-trailers (TT's) used for carrying containers to and from the ports, are not allowed to use containers in domestic trade. Neither swap bodies nor van-type semi-trailers are in use anywhere. Owners of "specials" therefore have limited scope to take advantage of the potential savings from using detachable trailers and hence offer shippers little incentive to consolidate their own loads.
(i) Inadequate freight-forwarding services make it difficult for TT's to achieve good utilization, bring down costs and offer shippers lower rates than 8- or II-tonners.
G) MOT's rates for TT's are high, so discourage occasional customers.
Possible economic cost
6.14 Larger trucks are appreciably less costly per ton-km, provided of course that they are fully loaded, or loaded as fully as the smaller alternative. Holding load factor, empty
41 A few exceptions are permitted under a grandfather clause.
27
running rate and annual mileage constant, Korean operators typically could save about 35% on vehicle operating costs by switching from a 2.5-ton truck to the next larger truck size available on the market, the 4.5-tonner; or 50-60% by switching to the largest rigid-body trucks or tractor-trailer combinations.
6.15 The removal of the three restrictions would lower the costs of owning and operating tractor-trailers, increase competition among TT owners, and transfer the incentive for making up full truckloads to the trucker or broker. TI operators would lower their rates and put pressure on operators of smaller trucks to lower their rates too. One might expect a two-tier pricing structure to emerge: one for full truckloads and another, higher per ton, for less-than-truckload shipments. Market information systems will have to be improved also; but the more competitive environment will give truckers a strong incentive in this direction.
6.16 Existing users of TIs would gain to the full extent ofproductivity improvements and narrowed profit margins. A tightening ofwaiting and loading times would reduce TT operating costs by about 5%; an increase in load factor on loaded trips, taking it up from its present somewhat low level nearer to that commonly found in unregulated countries, would cut another 5%; and the squeeze on profit margins might take the total reduction in rates to 15%. Such a cut in rates for existing customers ofTT's (10-15%) would be worth 12-18 billion won ($15-23 million).
6.17 The lowered rates and greater freedom to make innovations in service technology would draw traffic from smaller trucks, both own-account and commercial. There are three ways in which shipments now carried in two or more small trucks can be combined in one large commercial truck. A shipper can consolidate his own loads by building up inventory for a few days and dispatching less frequently. Alternatively the trucker consolidates loads from several shippers in a single truck travelling door-to-door but making extra stops to pick up and deliver a second (or possibly third) shipment. The third option is for shipments to be picked up separately in small trucks and consolidated at common-user terminals, as the "route" companies do already. The door-to-door option will cost less than the terminal option where access to the shippers' premises presents no difficulties for large trucks and just two or three closely placed pick -ups will fiII the truck. In the US shipments of over about 3 tons normally go door-to-door and smaller shipments go via terminals.
6.18 To evaluate the potential for shippers to consolidate their own loads by reducing dispatching frequency ideally requires disaggregate data on volumes to be distributed each week or month over specific routes. This is not available. When total movements are viewed in the aggregate, it is tempting to infer that there are large-scale opportunities for saving money by switching to larger trucks (para. 6.14 above). Indeed there are some grounds for concluding that many freight managers are insufficiently cost-conscious and unaware of ways now open to them to reduce their distribution costs. However, as a general rule we have to assume that shippers' present choice of truck size correctly reflects the prices they and their customers face and their valuation of time and quality differences, and it is these which lead them to ship often and in small lots.
28
6.19 A rough estimate of the potential benefits from shippers consolidating their own loads can be obtained by relating the quantity "diverted" to larger trucks, as a percentage of the existing divertable volume, to the percentage cut in price of the alternative "mode". EEC research on modal choicesl finds that cross-elasticities of demand relative to tariffs vary from well below -1 to above -5, the higher values applying to commodities that do not require special handling or special-purpose vehicles. Where the volume to be shipped is small, also, cross-elasticity is greatest; the shipper is more flexible in his choice ofmode. Assuming the elasticity is in the range 2-5 and the price change is 10-15%, the savings would be in the broad range 15-95 billion won ($20-120 million).
6.20 The second consolidation option, namely having a tractor-trailer pick up loads from more than one customer, is attractive only where both origin and destination ofthe separate shipments are close and their timing coincides. It is not normally feasible for bulk commodities or perishables. Using the simple model described in Appendix 2, we estimate that this option is cheaper than separate trips in 8- or II-ton trucks for hauls beyond 50-100 km, if the TT has to add no more than 15 km to its trip at each end for the second pick-up; the less extra driving needed, the lower the break-even distance. Assuming that the spatial and temporal distribution of movements is right for 25-50% ofthe potentially consolidatable commodities, the savings in supply cost would be about 20-40 billion won (US$ 25-50 million) per year.
6.21 The third option, consolidation at a terminal, entails several additional costs compared to the door-to-door movement: pick-up and delivery, unloading and reloading at the terminal, and storage and inventory carrying cost for the time it takes to build up a full truckload. The break-even distance, beyond which the terminal consolidation and line-haul in a TT becomes cheaper, depends on how far it is to the nearest terminal, the value of the commodity (and hence its inventory carrying cost), and the terminal's handling charges.
6.22 Again using the model described in Appendix 2, if all potentially consoli datable commodities now carried in 2.5-ton or 4.5-ton trucks required pick-Up and delivery trips totaling 20 km, and the inventory and handling costs totaled 3,000 won per ton (a middling estimate), then the break-even distance for small shipments would be about 150 km, and for medium-sized shipments about 200 km. It happens that halving or doubling the pick-up and delivery distance also halves or doubles the break-even distance. Raising the inventory and handling cost to 5,000 won per ton raises the break-even distance for small shipments to 200 km and for medium shipments to 300 km. These estimates also depend heavily on assumptions as to the value shippers place on the quality advantages of the handy, all-purpose small private truck and hence their willingness to move away from it.
6.23 There is therefore considerable uncertainty attached to the benefits ofterminal consolidation. Our best estimate is in the range 20-40 billion won ($25-50 million) per year. Five to ten million tons would be affected, up to double the volume at present moving by route trucks (4.6 million tons in 1983) and three to five times greater in terms of ton-km.
51 World Bank and Netherlands Institute of Transport, "The Determinants of Modal Choice in the Freight Transport Market", 1984.
29
6.24 The combined benefits from all these four effects would be 65-150 billion won ($80-190 million) per year. This is somewhat less than the simple sum of the effects taken individually, because of overlaps. Nonetheless it amounts to a reduction of 2-4.5% in total expenditures on trucking. The number of tractor-trailers would grow between 50 and 150%, taking their share of the fleet to as much as 8%.
6.25 As a consistency check, we assigned the consolidatable tonnage to small and large trucks in the same proportions as in the UK (though still using Korean costs). Making a reasonable allowance for the costs of consolidation, the savings were very similar (70-190 billion won, or $90-240 million).
Conclusion
6.26 The potential benefits are substantial, as are the changes in trucking techniques involved. The importance of this issue, and its complexity, have already been recognized by the MOT in launching the terminals study.
B. Operating Area Limitation
Background
6.27 Trucks with an "area" license are not allowed to carry a load for which both origin and destination are outside the truck's home province. In practice neither police nor competing truckers object when they see an operator violating this rule occasionally, but persistent offenders will be prosecuted, or harassed by the local truckers.
Hypothesis
6.28 The uarea" limitation requires more vehicles than would otherwise be needed, because it prevents seasonal migration from parts of the country where demand is slack to where it is at a peak.
Evidence
6.29 Prosecutions of uarea" truckers for violating the area restriction are rare, but the fact that they do occur indicates that there are geographical imbalances of supply and demand which could be eased if trucks had greater flexibility as to where they can operate. There is also anecdotal evidence of rates during harvest peaks being double what they are at other times ofyear.
Possible economic cost
6.30 Month-by-month truck traffic counts for 1983 from 15 points in seven of South Korea's nine provinces (Table 5) show that different provinces have different peak months: two in March, one in April, one in May, one in June, one in November and one in December. (Data are not available for the other two provinces.) Much ofthis can be attributed to the marked seasonal differences of the Korean climate, both as to temperature
30
and rainfall, so that, for example, crops are harvested earlier in the South than the North, and to provinces specializing in different crops. In five provinces the peak month is 13-20% above the annual average (after allowing for long-term trend growth), whereas the national peak month is only 6% above the annual average.
6.31 Both owners of trucks and provincial licensing officials decide on truck capacity levels with an eye to the peak demand, since temporary hiring in of spare capacity from other provinces for intra-provincial work is not permitted. If one makes the extreme assumption that the number of "area" trucks is strictly proportional to the truck ADT in the peak month, one fmds that, because of the complementarity ofthe peaks, 10% more trucks are needed, taking the seven provinces each as a separate stock rather than combined as a single pool. Since shipments to another province can be carried by trucks of that province, it is more realistic to assume that the truck fleet is set at 60-80% ofthe difference between the peak month and the annual average, i.e., the excess is 6-8%.
6.32 In practice truckers do not have perfect knowledge of where demand peaks are occurring and are not perfectly mobile. On the other hand, almost every point in Korea can be reached within one day from anywhere else (except Cheju Island), so lack ofmobility is not a severe constraint. One might reasonably expect 60-70% of the potential benefit to be realized, or a reduction of 3.5-5.5% in the number of trucks needed: 1,400-2,200 trucks scrapped each year would not need to be replaced. The savings would be their fixed costs, worth 15-20 billion won (US$19-25 million) per year.
31
TABLE 5: SEASONAL VARIATION IN TRUCK TRAFFIC, BY PROVINCE
Province J F M A M J J A S 0 N D Peak
Month Peak
% % of Area Truck Fleet
Kangwon 87 80 114 113 102 101 107 108 91 101 97 96 Mar 114 5.6
N. Chungchon 87 94 113 113 106 100 92 108 95 113 88 89 Mar 113 3.8
S. Chungchon 88 90 86 103 101 98 95 92 91 111 142 100 Nov 142 6.7
N. Cholla 70 86 102 120 108 101 97 94 107 105 109 101 Apr 120 4.6
S. Cholla 112 91 96 94 102 118 90 102 88 105 91 111 Jun 118 7.6
N. Kyongsang 85 82 100 93 119 108 99 108 101 103 97 104 May 119 12.3
S. Kyongsang 91 101 101 97 102 102 101 98 95 102 102 108 Dec 108 20.0
Kyonggi NA NA NA NA NA NA NA NA NA NA NA NA NA 38.l
Cheju NA NA NA NA NA NA NA NA NA NA NA NA NA 1.2
Total 90 91 101 101 106 104 98 101 95 104 103 104 May 106 100
Weighted average of peaks 117
Source: Trucking Regulation Study, Final Report, May 1985, pp. 197-212
32
6.33 Secondary effects would add to these savings. A modest reduction in empty running could be expected. The average empty running rate among "area" trucks excluding dmnp-trucks is now 38%. In the UK, where haul distances are similar, the average for public hauliers is 30%. A decrease of one percentage point in Korea would save the economy about 11 billion won (US$14 million). Ifmarket information systems were improved (brokers, freight forwarders, etc.) a reduction of 2-3 points would not be unreasonable. This would bring the total benefits to 30-50 billion won (US$45-65 million) per year, or 1-1.5% of total trucking costs.
6.34 A further, though probably smaller, effect would be felt by farmers and others who now own their own trucks to meet peak demand, for fear that commercial trucks will be scarce when they need them most. If free to hire out-of-province truckers, these farmers would have less need of such standby capacity. We have not attempted to quantifY this.
Conclusion
6.35 The complementarity of seasonal peaks is beyond doubt, but the extent to which truckers would migrate ifallowed to, can only be a matter of conjecture. The sole beneficiaries from the status quo are the railway, and truckers who are less aggressive in their marketing than their would-be competitors from other provinces. The losers are shippers and the more dynamic (yet law-abiding) truckers. The possible economic cost, while moderate, is large enough to justifY further consideration.
C. Ban on Own-account Truckers Operating for Hire
Hypothesis
6.36 The ban on own-account (OA) trucks operating for hire raises costs by inducing extra idle time.
Evidence
6.37 The large number ofprosecutions is strong evidence that this restriction is binding.
6.38 International comparisons also suggest that the own-account/commercial split is not optimal: that choices at the margin between OA and commercial carriage are distorted by the legal distinction drawn between them. The own-account share of the fleet in Korea (72%) is high relative to most countries with unregulated trucking systems. The OA share of the fleet in Brazil is 56%, in Chile 43%, in Peru 40%, in Kenya 52%, and in the UK 66%. Only in Thailand, among unregulated countries, is it known to be high (89%), but there are grounds for thinking that more costly registration procedures for commercial trucks deter owners from declaring them as such, so the Thai figure should be discounted. In heavily regulated systems own-account transport is often an escape valve, a way for shippers to avoid high-cost or low-quality commercial services. Thus in Portugal the OA share of the fleet is 90%, in Morocco 90% and in Tunisia 78%. (The OA share of truck traffic on the road, i.e., of total veh-kms, is always smaller than the fleet; the OA share of inter-urban traffic is smaller still, and ofton-km performed is yet smaller. This is because
33
commercial finns tend to specialize in larger trucks, and larger trucks operate primarily on interurban hauls and achieve higher annual mileage).
6.39 As to evidence of the effects of this regulation, there is only limited and conflicting evidence on the empty running rates and load factors among own-account trucks. The most reliable data, from the tenninal study's O-D survey, put the empty running rate for trucks other than dumpers at a respectable 31-34%. The load factors on loaded trips are reported(by the regulation study's company survey) to be consistently high (95-120%). The one perfonnance indicator on which they do appreciably worse than commercial trucks is annual mileage (32,000 km per year as an average for general-purpose trucks compared to 62,000km).
6.40 Some part ofthis difference is no doubt due to the private trucks specializing in activities where they have a comparative advantage. Just as small trucks are more often used for short hauls than long hauls (para. 6.12), so too are own-account trucks. Their advantage vis-a-vis commercial trucks lies in greater reliability and the provision of sales-related services by the driver, while their disadvantage is nonnally higher fixed operating cost per ton-km due to lower annual mileage. Their disadvantage weighs more heavily on longer hauls.
6.41 Several own-account operators reported that they own only as many trucks as they need to guarantee an essential minimum transport capacity, and use them whenever they can be assured of a backhaul, such as delivering the goods they sell on the outbound trip and bringing back components and supplies on the return. When no backhaul is in prospect, they hire a commercial truck. If they were pennitted to carry freight for others on backhauls, they would make greater use of their trucks, thereby raising their annual mileage.
Possible economic cost
6.42 In the extreme event that all general-purpose private trucks (that is, excluding dwnp-trucks, tankers and tractor-trailers, which in Korea are mostly special-purpose vehicles) raised their annual mileage to that of commercial trucks, the effect would be to raise the total output of private trucks (over I ton) by about 22%, saving 280 billion won per year in fixed costs of trucks that would no longer be needed. In practice such a large response is unlikely. In the UK, after commercial hauling by own-account trucks was deregulated, such work accounted for only 4-6% of their activity_ If the same level were to apply in Korea, the saving vis-A-vis the status quo would be in the range 50-75 billion won (US$ 65-95 million), or 1.5-2% of total expenditures on trucking.
Conclusion
6.43 Many governments, influenced by the lobbying of the powerful commercial trucking profession, are strongly against commercial activity by own-account trucks. Those who
34
stand to lose are, of course, the commercial carriers. The arguments they often put forward are that such activities (a) destabilize the market, (b) are a danger to shippers because private truckers are unsafe or uninsured, and (c) rob the Treasury of tax revenues. To the first argument it can be said that trucking has none of the characteristics that make an industry inherently unstable (in contrast, for example, to civil aviation): the fixed assets (trucks) can be purchased in increments that are not large relative to the variable costs of operation (unlike aircraft and airports); lead times to add to or redesign these assets in response to a surge in demand are not long (a matter of weeks for trucks compared to years or even decades for planes); the financial life of a truck is not long either (they are normally written off in ten years or less); and the assets of a company that fails can readily be redeployed with little loss to the economy (much less than aviation ground facilities).
6.44 As for the other arguments, all trucks should have to meet safety and insurance standards, regardless oftheir ownership status, and taxation ought to be put on a basis that does not discriminate between private and commercial owners.
6.45 The magnitude of the probable cost of this regulation is moderate. The potential beneficiaries from deregulation are shippers, normally dispersed and not vocal. The issue is worth exploring further.
D. Quantity Restrictions on Yongdals
Background
6.46 Yongdallicenses command a resale value, implying that license authorizations are not enough to satisfy the demand.
Hypothesis
6.47 The entry restrictions bring down yongdal service quality and create economic rents for operators, which are passed on in higher prices.
Evidence
6.48 The resale price for a yongdallicense is commonly 10 million won in Seoul (US$ 12,500). In Kwangju, a provincial capital, it is reported to be 8 million won ($10,000) and in Taejon and Chongju, other provincial capitals, 4-5 million won ($5,000-6,000). One would expect the various prices to reflect a greater or lesser shortage of supply. However, we have failed to find a statistically strong explanatory relationship. Seoul's high premia could be explained by the fact that it has the fewest yongdals in proportion to its popUlation or zonal income and experienced the least growth in yongdal numbers between 1980 and 1984. But Kwangju, which has the highest yongdal density and where their numbers have grown rapidly, still has high premia; while Chongju has low premia but also low yongdal density and slow growth.
6.49 The study's authors were able to substantiate that some applications to purchase new yongdals in Taejon were refused in 1985. Other telling evidence that the entry
35
restrictions are binding is that yongdal operators commonly demand considerably more than the meter rate, and are described by shippers as·hard to find when they need one.
Possible economic cost
6.50 No reliable information is available on the profit margins of yongdals. The annual operating cost of the typical yongdal is about 8.5 million won (US$ 10,000). The real interest cost on the amount "invested" in the license, at 10% p.a. (allowing for a substantial risk premium on top of the risk-free real rate of interest), will be 1 million won in the case of Seoul, or 12% ofVOC, and down to 5-6% ofVOC elsewhere. The owner presumably recovers this through a mark-up in the prices which he negotiates with the customers. The welfare loss to the economy (assuming that their price elasticity of demand is 0.3-0.5, which is probable) on the total yongdal revenues of about 150 billion won ($180 million) will be something less than 1 billion won ($1 million). On the other, if there were no quantity restrictions on yongdals, customers might well be willing to pay more for greater availability, i.e., a superior service quality (by analogy with taxis). As a rough guess we value this at 10% of the present revenues, i.e., 15 billion won (USS 18 million).
Conclusion
6.51 The premia are large relative to the purchase price of the vehicle (as much as double in Seoul's case), while the welfare loss to customers on account of the deficient service quality is small relative to total trucking costs, because the yongdal's share of the total market is itself small.
E. Ji-ip SystemIMinimum Fleet-Size Regulation
Background
6.52 Individual owner-operators have seen required by law to be affiliated with licensed commercial trucking companies to meet the minimum fleet size regulation. They pay fees to the companies of0.8-1.3 million won ($1,000-1,700) per year, which they allege far exceed the value of the minimal services (e.g., processing insurance and licensing paperwork) they receive in return. Since 1985 those with trucks under 5 tons capacity are now being released from this obligation over a four-year transition period.
Hypothesis
6.53 The ji-ip fees are economic rents which cross-subsidize inefficiency among company-owned trucks.
Evidence
36
6.54 The fact that three out of every four owner-operators allowed to do so are now opting out of the ji-ip system is convincing evidence that the restriction was binding. As to its effects, the hypothesis suggests that the more "satellite" owner-operators a company has, the fatter and lazier it becomes. One would expect this to manifest itself in higher overheads among fleets with a large ji-ip share, or higher direct operating costs because of more idle time, empty running or higher payments to labor. Company financial data collected by the survey provide only weak evidence in support of this hypothesis. Regression analysis of direct operating costs of company-owned trucks against the company-owned share ofeach fleet shows a negative slope, consistent with the hypothesis, for all truck sizes except the largest (see Figure 2). However, the correlation is very weak: not higher than R2 = 0.19, which is not statistically significant. The regressions stand or fall on the strength of only two or three outlying firms with 0-50% owner-operator trucks. Any conclusion needs to take account of the very wide variation in unit cost among firms with similar ji-ip shares (80-90% ji-ip, 10-20% company). The variation is as large as 1 to 4, suggesting that some firms specialize in short-haul business (the high-cost operators) and others in long hauls. Geographical segmentation of the market or specialization by commodity could also contribute. The data needed to standardize for such segmentation are not available.
Possible economic cost
6.55 The economic cost of this regulation has a small direct component and a large indirect one. About 86% of commercial trucks over I-ton capacity are ji-ips, i.e. about 38,000. The total ji-ip fees are therefore about 40 billion won ($45 million) per year, on revenues of about 760 billion won ($950 million). Assuming that the cost of the services to the owner-operators is about one third of the revenues fromji-ip fees, the rent element is about 25 billion won ($30 million). The welfare loss to shippers, again assuming that their price elasticity of demand is 0.3-0.5, is less than 1 billion won ($1 million). This is hardly significant on its own.
6.56 On the other hand, it can be argued that, if the company-managed fleets were not receiving the ji-ip fees, which amount to 15-20% of the revenues on their own trucks, they would have to operate more efficiently to survive. That is, they would be forced by competition to make cost savings of the same order of magnitude, i.e. about 25 billion won ($30 million) per year (approaching 1% of total truckinKcosts).
Conclusion
6.57 The question whether larger fleets offer economies of scale is evaluated in Appendix 3. The present hypothesis concerns the effect of the ji-ip/company relationship and the ji-ip fees. The evidence, such as it is, supports our hypothesis, but more analysis of the explanation for the wide variation in cost levels among firms would be welcome. The possible economic cost is moderate.
37
Figure 2 TRUCK DIRECT OPERATING COSTS
IN RELATION TO OWNER-OPERATORS' SHARE OF COMPANY FLEET
WON
800
x
x
x
x
x X
X
X
x
o 20 40 60 80 100 % WON OWNER-OPERATOR TRUCKS AS % OF COMPANY FLEET
~ (/J
o~
$~;:2u l.I.l~>0:::of-< .....lOUl.I.l
g~f-<O(/J ,
O~U<c ~c... ~::;sf-<O
~~ l.I.lO
~~ uo::: ~~ ....
o 20 40 60 80 100% OWNER-OPERATOR TRUCKS AS % OF COMPANY FLEET
800T T
600
400
200
8 TON TRUCKS
X X
R2=O.19 X
X X
X
0
39
Chapter 7: CONCLUSIONS AND RECOMMENDATIONS
Conclusions
7.1 The estimates ofthe possible economic cost of the regulations examined in the previous chapter may be summarized as follows:
1. Ban on part-loading, use of containers domestically, etc.
2. "Area" limitation
3. Ban on private truck.., operating commercially
4. Y ongdal entry restrictions
5. Ji-ip fees
Total
Billion won
65-150
35-50
50-75
15
25
190-315
US$ million
80-190
45-65
65-95
20
30
240400
% of total trucking expend.
1.9-4.4
1.0-1.5
1.5-2.2
0.5
0.7
The totals and conversions are rounded to the nearest 5, to avoid spurious impressions of accuracy. The combined effect may not be strictly additive, but again, these are no more than very rough estimates.
7.2 They serve nonetheless to show that the benefits of regulatory reform in trucking would indeed be large: US$ 240-400 million per year. They also show that the highest priority should be attached to the truck size/consolidation issue. Furthermore, they suggest that lifting the ban on private trucks operating for hire, though hotly opposed by the commercial trucking lobby, might nonetheless be the second most beneficial among the reforms examined for the economy at large. Essentially this is because it affects four-fifths ofthe total truck fleet and two-thirds of the trucking market in cost terms.
7.3 With regard to the truck size question, it is significant that the scope for freight consolidation and the use of larger trucks is no greater. Our modeling indicates that the share of tractor-trailers in the fleet may well double or more, taking it from 3% of trucks over 1 ton to 5-10%. But this growth will come at the expense ofmiddle-size trucks as much as of small ones, and the small trucks' share will remain high. L TL terminal services will capture no more than perhaps 10% of the road freight market (in terms ofton-km). Much can be explained by the shortness of haul distances. The findings of our modeling
40
coincide with observations in other countries in concluding that the break-even distance for commercial freight consolidation services is in the range 150-250 lan, and in Korea's case, that excludes 60-80% of all movements of commodities suitable for consolidation.
Recommendations to the Korean Government
7.4 We therefore recommend that the Government:
(1) Eliminate the distinction between area, route and special licenses (with the effect of allowing ex-area trucks to take out-of-province backhauls and ex-area and ex-specials to part-load).
(2) Limit operator licensing criteria to safety and professional competence.
(3) Give licensed operators the freedom to choose the number of trucks they operate and their size.
(4) Permit the use of containers in domestic trade.
(5) Do not restrict private trucks as to quantity, and let them backhaul for hire, subject to the same safety and competence requirements as commercial carners.
(6) Strengthen safety enforcement mechanisms.
(7) As regards truck taxation, remove discrimination by type of ownership.
(8) Extend the new policy of allowing certain owner-operators to work independently to all, regardless of truck size.
(9) Leave brokers alone.
(10) Monitor the productivity and cost indicators listed in Chapter 5 by sample surveys to see whether the predicted effects materialize and to provide a sounder basis for evaluating government intervention in the trucking market. (Detailed proposals are given in Appendix 4.)
41
Table Al TRUCKING OUTPUT: FLEET COMPOSITION AND VEHICLE KMS
2 3 4 6 7 8 9 10 I 12 13 14 15 16 2+4 2x8 4x9 11+13
ANNUALKM NUMBER OF TRUCKS PER TRUCK TOTAL VEHICLE-KMS PER YEAR
Truck Capac: Commercial Private Total Comm. Priv. Avg. • Commercial Private TOlaJ _]JP: ___________ !~P!_~____ {~t __ J~l _____ _ ____ <r~L _____ j9QO)____ (01l __ ~ ____ (092L _!QOPl __ !~! ____:___(.fT!!1L ____(YO) ____~2 _____ 1'&1. ___(!1'!.n) ____ (~l. __ __ J
: Yongdal : Y ongdal • Y ongdaJ :
Pick-up I. ° 17.6 29 114.9 50 132.5 45 32.0 25.0 25.9 563 17 2,873 44 3,436 35 : Area Area Area
Standard (S) 2.5 11.7 19 83.4 36 95.1 33 36.4 25.6 26.9 426 13 2,135 33 2,561 26 4.5 5.6 9 17.6 8 23.2 8 70.5 32.1 41.4 395 12 565 9 960 10 8.0 g.1 \3 2.0 10.\ 3 79.4 71.9 77.9 643 19 144 2 787 8
11.0 4.0 6 1.2 5.2 2 77.9 67.3 75.5 312 9 81 1 392 4 Subtotal S 29.4 48 104.2 45 \33.6 46 60.4 28. I 35.2 1,775 54 2,925 45 4,700 48
Dump (D) 4.5 \.8 3 3.8 2 5.6 2 61.9 56.6 58.3 III 3 215 3 327 3 8.0 5.9 10 2.9 I 8.g 3 61.9 56.6 60.2 365 11 164 3 529 5
11.0 1.6 3 \.2 I 2.8 I 61.9 56.6 59.6 99 3 68 1 167 2 Subtotal S+D 38.7 63 112.\ 49 150.8 52 60.8 3~.] 37.9 2,351 71 3,372 51 5,723 58
Route Route Route Route 8.0 1.3 2 1.3 o 59.7 78 2 78
Special Special Special Tanker 10.0 0.6 2.7 I 3.3 J 88.5 88.5 88.5 53 2 239 4 292 3 Tractor-t 23.0 3.5 6 0.9 o 4.4 2 78.5 80.0 78.4 273 8 72 I 345 3
Tolal >1 Ion 44.1 71 115.7 50 159.8 55 62.5 31.8 40.3 2,755 83 3,683 56 6,437 65
Grand totallAvg. 61.7 100 230.6 100 292.3 100 53.8 28.4 33.8 3,318 100 6,555 100 9,873 100
CorrnnIPriy % 21 79 100 34 66 100
Trailer 5.5 2.0 7.5
Sources: Truck numbers: MOT Annual Kms: Trucking Regulation Study, except for dump trucks and
tankers, which are taken from Road User Charges
42
Table Al (b) TRUCKING OUTPUT: LOAD FACTORS AND TOTAL TON-KMS
17 18 19 21 22 23 25 26 27 28 29 20 24 30
1x17 18x(1- : 1x21 I Ix20 13x24 19) : 22x(l-23) : 25x27
COMMERCIAL : PRIVATE Truck . Load Avg. Empty A v : Load Avg Empty Av: TOTAL TON-KMPER YEAR Capac : Tons tons : Commercial Private Type : Factor Load Run Factor Load Run : Total ,,(ton) : perkm perkm (%) (mn) (%) (mn)
: (%) (tons) : (%) (tons) (%) : Traveled : Travld ,
.---------- ---------;-------- ------)io~g~l----------------:--------------------------- - ----:--)To~!id~i-- ----------------------------------, Pick-up : 73 0.7 45 : 73 0.7 45 : 226 2 1,149 10 1.0 : 0.4 0.4 : 1,375 6 , . , .. .
: Area : Area Standard ,: 112 2.8 43 96 2.4 32 .: 680 5 3,484 31 2.5 : 0.6 1.6 : 4,164 17, . (S) : 111 5.0 38 : 102 4.6 31 : 1,224 9 1,793 16 4.5 : 3.1 : 3.2 : 3,017 12
8. 9 33 : 106 8.5 34 : 3,835 29 807 7 8.0 : 5.6 : 4,642 19
11 .4 40 : 118 13.0 34 : 2,131 16 693 6 11.0 : 8.6 : 2,824 11
Subtotal 7.1 38 : 3.4 32 : 7,870 59 6,777 60 4.4 2.3 : 14,647 59
I
Dump 100 4.5 54 : 100 4.5 54 : 231 2 445 4 4.5 2.1 : 2.1 : 676 3 (D) 100 8.0 54 : 100 8.0 54 : 1,344 10 604 5
43
8.0 3.7 : 3.7 : 1,948 8 I100 11.0 54 : 100 11.0 54 I
501 4 344 3I
I11.0 5.1 :5.1 I
845 3I
Subtotal S+D 7.3 42 3.7 35 : 9,946 74 8,170 73 4.2 2.4 : 18,116 74
I Route Route
Route 79 6.3 10 440 3 8.0 5.7 Special
Special 80 8.0 41 244 2 1,128 Tanker 90 9.0 49 : 4.7 1,372 6 10.0 4.6 : 78 17.9 41 : 2,562 19 762 7I
Tracktor-t 60 13.8 32 : 106 : 3,324 13I •
I I23.0 9.4 , I I I I, I I I I
Total >1 ton 0.8 40 4.2 35 :13,191 98 10,060 90 4.8 2.7 :23,251 94I
I IGrand total! A vg I
I
I
I
6.8 41 2.8 40 :13,417 100 11,209 100 CommlPriv % 4.0 1.7 :24,627 100
54 46 100
Sources: Load factors: Trucking Regulation Study Empty running: Freight Tenninals Study
44
Table Al (c) TRUCKING OUTPUT: COSTS PER TRUCK AND FLEET TOTAL COSTS
31 32 33 34 35 36 37 38 39 40
UNITV.O.C. Variab Fixed COST PER T-K ANNUAL SUPPLY COST
Truck Capac Perkm per yr Cornm. Priv. Commercial Private Total ___________~~~>- __ __ 1___ _("!~tY_____ !Q~ "Yl. ____ .!~!?!:l _____ J~_0!l2 __ : __ {~n)_____ __. _____~l _____ ~~t _______
Pick-up 1.0 NA NA 650 700 147 14 804 36 951 29
Area Standard (S) 2.5 95 7,945 196 248 133 13 865 39 999 31
4.5 119 8,335 77 119 94 9 214 10 308 9 8.0 178 9,719 50 56 193 19 45 2 238 7
11.0 215 11,332 53 45 112 II 31 I 143 4 Subtotal S 533 51 1,155 52 1,688 52
4.5 119 8,335 123 129 28 3 57 3 86 3 8.0 178 9,719 91 95 122 12 57 3 180 6 1.0 215 1l,332 79 82 39 4 28 1 68 2
Subtotal S+D 723 70 1,298 59 2,021 62 Route
Route 8.0 178 9,719 60 26 3 26 Special
Tanker 10.0 215 11,332 75 73 18 2 82 4 100 3 Tractor-t 23.0 266 13,895 47 42 128 12 32 153 5
Total >1 ton 67 140 889 86 1,412 64 2,300 71
Grand totall A vg 77 198 1,036 100 2,216 100 3,252 100
CommlPriv % 32 68 100
Source: Unit V.O.c.: Trucking Regulation Study
45
Appendix I
DESCRIPTION OF SURVEYS CONDUCTED
1. The consultants conducted three surveys: (a) a questionnaire sent to commercial truck companies, (b) a questionnaire sent to owners of private fleets, and (c) a roadside survey. They were carried out in November and December of 1984.
Commercial company survey
2. The company questionnaires asked for data on (among other things):
• license category • fleet size
for each truck:
• ownership (company or owner-operator)
• SIze
• age
• distance traveled so far
• distance run in 1984, loaded and empty
for the company:
• monthly volume ofbusiness • commodities carried • number statement • itemized vehicle operating expenses and overheads • route map ("route" companies only)
3. The questionnaire also asked for opinions on the option of eliminating the distinct area, special and route licenses and the carrier's likely reaction to such a change if it were made.
4. The commercial company survey initially set out to sample 10% of the over 1,200 "area" and "special" fleets and a1125 route companies. The questionnaire was distributed through the provincial truck associations, who each called a meeting of the sample companies at which the purpose of the study was explained and general questions were answered. The companies were each contacted by phone and asked to return the questionnaire by mail.
5. A total of 102 questionnaires were returned, a response rate of 47%. Because of shortage of time and the late return of some questionnaires, 16 were not analyzed. This left 86, giving the following sample statistics:
46
Questionnaire Questionnaire %of Retumed& %of
POQulation Distribution Pop. Analyzed Pop. ComQanies Area 864 125 14 49 6 Special 344 66 19 22 6 Route 25 25 100 15 60
Total 1,233 216 18 86 7
Trucks 2.5 ton 11,700 542 4.6 4.5 ton 5,700 494 8.7 8 ton 11,800 825 7.0 10 ton or over 11,200 162 1.4 Tractor-trailer 3,300 133 4.0
Total 43,700 2,156 4.9
6. Trucks under 1 ton were not included, since the consultants' terms of reference did not cover yongdals. In the event, summary data on them were obtained from the Seoul Yongdal Association.
7. The sample was inadvertently skewed towards companies of below-average size, because companies in Seoul and Kyonggi province (the one surrounding Seoul) tend to be larger and it was in these two areas that the questionnaires were distributed last and the response rate was lowest. Whether this introduced bias into the data is not known; the consultants had no prior basis on which to make adjustments in calculating population averages for costs and productivity indicators.
8. The sample was also skewed as to truck size: 4.5-ton and 8-ton trucks were well represented and over 10-ton trucks only poorly. Again the consultants had no prior basis for making adjustments in the cost and productivity averages on this account.
9. The reliability of the financial data was jeopardized by the fact that many questions were designed to match tax returns or to conform to MOT regulations. This will have biased the responses towards what the company thought they ought to be, rather than what they really were.
Private fleet survey
10. The private survey form was intended to produce operational data and qualitative information on the reasons private operators gave for using their own trucks rather than commercial trucks. Since private companies were thought not to keep proper cost records of their trucking operations, they were not asked for financial data.
47
11. Private truck fleets are owned by manufacturing finns (about 39,000) and distributors, retailers, builders and others (about 210,000). Of the latter, more than half (119,000) are under 1 ton capacity and believed to be owned by retailers or individual traders. The selection of a sample was not random, but done on the basis of suggestions from the provincial truck associations. Forty finns, evenly distributed among the provinces, were sent questionnaires; 25 returned them (a 63% response rate). The returns covered 1,500 trucks over I ton capacity, or under 1 % of all private trucks.
12. The sample was skewed towards smaller finns and the less industrialized provinces. Private trucks are concentrated in Seoul, Pusan and South Kyongsang (the south-east), where half of all private trucks are registered. Because of the equal distribution by province, however, only 30% of companies surveyed came from these locations. Nonetheless, in the fonns returned there was little difference between the perfonnance of trucks from Seoul and Pusan and those from other provInces.
Roadside survey
13. The aim of the roadside survey was to provide cross checks on the company surveys as to perfonnance indicators among trucks actually moving on the Korean road system: age, total mileage, empty running and part-loading by truck size and ownership type.
14 Difficulties in organizing the survey and lack of budget forced the number of survey points down and the locations to be mostly expressway rest areas and tollgates. Each point was surveyed for 24 hours; all were surveyed in one week in December. The roadside survey was therefore basically a sample of expressway running, supplemented by other cordon line surveys at three major cities conducted for other purposes.
Supplementary surveys
15. Supplementary surveys were conducted of brokers and other economic actors, andji-ip trucks. Ninety ji-ip operators were interviewed at expressway rest areas.
Seasonality
16. How the running data reported by the companies for November 1984 should be factored up to a yearly total depends on the nonnal pattern of monthly variation where that month stood in the business cycle and trend growth. A separate survey showed that industrial shipments in November 1984 were up 18% on the previous November and 9% over January. Ministry of Construction monthly traffic counts show that in 1983 travel volumes in November were above the annual average at most locations, by 10-20% on expressways and 5-10% on national roads. The seasonal variation can only be established by comparing data from several years, which we do not have. Trend growth could explain about 4% ofthis variation, but was offset by the fact that the economy was slowing in November 1984. It is therefore possible that multiplying November 184 output by 12 overstates the annual total by perhaps 5%. Because of the softness of the evidence, however, no such adjustment has been made in the perfonnance indicators presented in this report.
48
Appendix 2
DETAILS OF COST CALCULATIONS
Truck Productivity and Consolidation Model
1. Data collected in the course of the regulation study and the freight terminals study make it possible to model the changes in truck productivity and size distribution that might be expected from the introduction ofpart-loading by commercial trucks, common-user consolidation terminals, and improved market information systems.
2. The base case, shown in Table A2, starts from a matrix ofexisting freight movements by road broken down simultaneously by distance range and truck size. The volumes were derived from an origin-and-destination survey carried out under the terminals study in 1985. They refer to a 65x65 -zone matrix and therefore exclude intra-urban trips. Also excluded are bulk commodities moving by road (coal, lime, petroleum products, iron ore, sand and gravel, and bulk cement) and other commodities that, by their nature, would not lend themselves to consolidation with other commodities (farm, forest and marine products). This leaves about 60% of total freight movements by road in the matrix. Regrettably it does not separate out private and commercial trucks.
3. A truck performance model was developed within the present study, simulating as far as possible the operating decisions facing truckers. Korean trucks typically are in use six days a week with Sunday off, and are used regularly over the same or similar distances. Thus, for example, trucks hauling from Seoul to Pusan (about 420 km) usually perform three round trips per week; they take eight hours on the road each way. Trucks on a short haul such as Seoul to Inchon (30 km) make one or at most two round trips per day; they spend 1-1.5 hours in transit each way. On short- and long-haul business alike, the rest of the day is spent in loading and unloading (one to two hours for each) and waiting, either in search of a load or because ofrestrictions on hours when trucks may drive in cities. Drivers work very flexible hours from one working day to the next, driving or sleeping by night or by day according to what is convenient; but they will make sure to get home for their one day off per week, i.e., there will be a whole number of round trips per week.
4. A speed equation was fitted to the known times in transit, as follows:
v 55 - 1000 d
where v average speed, origin to destination (in km/hr) d haul distance (in km)
5. It is known from the two studies cited that 31-43% of all inter-urban running is empty; i.e. trucks secure back-hauls on 15-40% of all trips.
--- ------------------
TRUCK CAPACITY
-~------~---
(TONS) 2.5 4.5
8-11 IT
TOTAL--_ ...... _-----
2.5 4.5
8-11 IT
TOTAL
AVHAUL (KMS)
2.5 4.5
8-1I IT
TOTAL
49
Table A2, BASE CASE
DISTANCE RANGE (Ians) -50 -100 -150 -200 -250 -300 -350 -400 -450 TOTAL
TONS (000)
12,860 6,823 3,405 2,562 1,604 599 697 436 36 29,022 4,758 5,244 2,943 1,450 1,920 724 787 652 34 18,512 7,719 9,496 5,615 5,815 553 3,403 3,773 6,261 781 48,416 5,868 6,935 2,507 1,805 2,509 1,150 2,403 2,689 305 26,168
_~!.3Q~ ___ ?~,~?~______ !~,~?Q _____ ]l,~??____ !!,?~~ _____5-,~?~ ____~,~ ___ _lQ,P~~ _____ !,!?~ __ -!~~,!!~-SHARE OF DISTANCE MARKET (%)
41 24 24 22 14 10 9 4 24 3 15 15 18 20 12 17 12 10 26 40 3 21
....25- -3.L -..39.-.. --5Q. --4l!. ----5&. ------49.. ----62 68 19 24 26
100 100 100 .. _........ _-----
TON-KMS (MN) 30 70 420
386 478 15 143 367
---rzr232 665 176 485 128
17 15 22 20 31 27
100 100 100 100 100 100
----_ ... _-------- -- ----------------------------------------
120 170 220 270 320 370
2,622 409 436 353 162 223 161 2,235
8,551 353 247 422 195 252 241 4,023
674 989 1,222 919 1,207 2317 328 17,431 301 306 552 211 769 995
486
100
-----------------------------
------------
50
HRSUSE
AVGSPD HRSffRP
EMPTY % CYCLIWK
EXC.TT KMSIYR EXC.TT
............ __ ... __ ....
2.5 4.5
8-11 IT
2.5 4.5
8-11
-_ .. _-------
2.5 4.5
8-11 IT
TOTAL
PERFORMANCE MODEL
PERWK 0.90
72 A V AILABIUTY
WAIT+LOAD+UNLOAD HRSffRIP
0.80
6 4
22 41 47 53
1.38 1.72 2.57 7.98
45 37 35 32
8 7 6 3 10 9 7
3 20 41 60 105 28 59 79 118
~ .. --------------------V.O.C. (WON} V ARIKM FIXIYR (000)
95 7,945 119 8,335 1% 10,486 266 13,895
COST PER TON-KM (WON}
320 164 124 90 195 103 79
58 128 70 55
43 86 49 42
35 COST PER YR (BN WON)
123 78 51 I
28 38 28
-JO- --4;L --3't 14 15 24 13
4
1% 187 129 21
49 50 51
3.46 4.36 5.26
33 33 33
5 4 4
6 5 4
71 73 90
95 103 101
AVGKMS
III 108
71 69
51 50
38 37
- -------_ .. _-_ ... _
48 38
18 29
--50. ----6f)
12 20
128 149
(000) 46 55 68 84
..... -_ ..
97
63
45
37
.. _.. _-_ .. _...
16
12
--4Z
12
81
EXC. IT
EXC. TT
52 52
5.17 7.07
32 32
3 3
4 3
80 92
120 104
EMPTY % TRUCK MIX %
36 48 35 18 33 27 34 7 -- -- --_ .. ----- ... -- --- -_ ...... --- - ... - .............
103 95
66 62
47 45
35 37
_.. _----------------
23 15
17 15
--57 -tffi
27 36
124 170
- ..............
395 185 441 163
1,184
---------~------.-- ...~.------- ... ---------------------- -------------------------- ... ---------
---
_ __ _
TRUCK CAPACITY
LOAD 2.80 5.00
10.20 16.10
TOTAL
LOAD 2.80 5.00
10.20 16.10
TOTAL
2.5 4.5
8-1I IT
2,5 4.5
8-11 IT
__ ... _J.(tf~~_
2,5 4,5
8-11 TT
2.5 4.5
8-11 IT
2.5 4.5
8-11 IT-_ .. __ .. __ .. -
51
Table A2, page 2 BASE CASE (continued)
DlSTANCE RANGE (kms) . _. :~q ___ ..-} 99_. __ .:1..5.9.. ___ :~C!.O. ___ . :~~~ __ . :~9Q_. LOADED VEH-KMS (MN)
138 17l 5
29 73 3 ~ ~
32 II 30 8
200 339
---m52 41 20
2"tl 117 103 48
364
3.52 3.53 3.14 6.29 6.30 5.61
12.82 12.85 11.45 25.30 26.08 21.09
NO, OF TRUCKS NEEDED 12,512 2,592 2,062
706
_F.1~~~
41 75
146
71
--66
19
301
156 126
49 84
---il1- -----UO
19 34
321 365
----m- ----zJ.r ---rn108 74 126 101 146 179
29 29 51
460 482 543 ----~----~-~-~-~--------------
TONS PER TRUCK PER DAY (AVG LOAD X LOADED TRIPS PER DAY)
6,623 3,709 2,851 1,795 2,530 1,679
809 362
___ !~,~!~ __ .. ?t)~?
58
39
~
19
206
~
58 134
29
306
2.66 2.15 2.16 1.62 1.63 1.63 4.75 3.83 3.85 2.90 2.91 2.91
9.69 7.82 7.86 5.92 5.93 5.95 18.36 15.43 12.41 12.45 9.37 9.38
(TONS/365ffONS PER TRUCK DAYIAVAIL %) 3,297 2,559 950 1,045 1,715 643 2,054 2,432 1,485
299 495 282
___ ~,§~?_. _. J2~C!.O____ !r3.5.!l..
1,469 929
2,183 587
917 768
3,614 874
76 40
450 99
~\I§~... _6A~~ .. ~..
39
(000 WON PER FULL TRUCKLOAD) 44 56 65 74 83 93 104 52 65 77 89 100 113 126
66 90 116 139 153 169 186 205 1I8 156 203 243 268 294 325 345
(000 WON PER AVG TRUCKLOAD) 42 53 67 73 47 61 76 85 68 88 III 125 81 105 131 162
RATE-TO-COST RATIO 0,79 0,84
1.05 1.80
1.02 1.05 1.05 0.98 1.00 1.07 1.09 l.03 om 1.05 1.32 1.32 1.32 1.25 1.22 2.12 1.93 1.93 1.85 1.66 ---------------------~~~- ... -- .. -----.
_.:~~~ ____ :4.0.°__ . ___ ::!~~
80 58
50 48
~ ~7
48 62
2% 395
---rT8 -----gs 74 7I
175 335 71 91
438 582
48
----g 4
47 12
71 -~--~--------~-
92 106 155 181
99 114 169 218
105 123 183 237
0,90 0,95 1.09 1.63
0.94 0,99
1.11 1.49
0,98 1.03 1.12 1.45
~------------------
TOTAL
936 447 838 250
2,471
1,462 684
1,260 378
3,784
32,112 12,378 18,486 4,513
.... _g~~~_
-. --_ ...... _..
52
Evidence from other countries indicates that the longer the trip, the lower the empty running rate. The time and cost of searching for a return load is invariate with trip distance, while the revenue is proportional to distance. Thus few truckers on trips only 30 km from home will take the trouble to find a return load, but many more will do so when they are 400 km out. The model assumed a 45% empty rate on 30 km trips and 32% on 420-km trips, with interpolation as an inverse function of distance in between; this gives an average, weighted over all distances, that calibrates with observed data.
6. The number of round trips per week and the average haul distance, together with the amount of down-time for maintenance and repairs, determine the annual mileage. The latter is known for different truck sizes. Down-time is known to be about 20% for all sizes except TT's, for which it is about 10%. The number ofhours in use per week (i.e. when both truck and driver are available) and the miscellaneous waiting time preceding each trip are not known, but were estimated by iteration so as to yield the known annual mileage. The best fit was achieved with 72 hours in use per week (e.g .. 12 hours per day for six days) and 6 hours spent waiting, loading and unloading on each trip. Tractors that use interchangeable semi -trailers do not have to wait for loading and unloading; the average waiting/loading/unloading time for them was estimated at 4 hours.
7. Truckers spend less time in miscellaneous waiting on a back-haul than a main haul; if the probability offmding a return load is low, they will return empty without delay. Also minor maintenance will be performed at home. Therefore, the average time for waiting, loading and unloading on a back haul was set at 70% of that on the main hauL
8. The parameters of the model described so far do not differentiate by truck size, except the shorter waiting time and down time for TI's. This is nonetheless consistent with larger trucks performing higher annual mileage, since the smaller trucks are used predominantly for short hauls while the larger trucks are active in both short and long hauls. Thus:
Truck size Avg. haul KIns per year (km)
Small (2.5 ton) 90 46,000 Medium (4.5 ton) 125 55,000 Large (8-11 ton) 180 68,000 TT (23 ton) 155 84,000
9. The regulation study estimated truck operating costs synthetically as follows (values for 1983):
53
Variable cost Fixed cost perkm per year
won US cents mn won $000
2.5 - ton 95 12 7.9 9.9 4.5 - ton 119 15 8.3 10.4 8-11 - ton 196 25 10.5 13.1 23 - ton 266 33 13.9 17.4
10. A separate cost per ton-km is calculated for each truck size and distance range. The fixed cost is allocated according to the annual krns that would be performed if the truck plied exclusively over the same haul distance throughout the year.
11. The average loading assumed is that reported in the regulation study for loaded trips:
Truck capacity Avg. load on loaded trips (tons) (tons)
2.5 2.8 4.5 5.0 8-11 10.2 23 16.1
12. These load factors, in the range 100-110% for all but TT's, are unusually high by international comparisons. However, the data, reported by truck companies trip by trip on sample days, is ofacceptable quality and can be explained by the adaptation of shippers' behavior to the requirement that they pay a commercial trucker not by the ton or cubic meter but by the full truckload. (The one exception is route business, where lower load factors are reported). Accordingly, they match consignment sizes to the available truck sizes. As for own-account operators, a number interviewed asserted that they use their own trucks whenever they can fill them and/or obtain a return load but hire a commercial truck whenever they cannot. This would explain why they too have high load factors on loaded trips and actually less little empty running than commercial trucks. Estimates ofload factors from the terminal study's O-D survey imply far less running full. However, they can be doubted on the grounds that only the surveyor's visual assessment was recorded --e.g. the truck looked half full-- without his knowing how much the load actually weighed. The only weighing stations are on the expressways and many smaller trucks do not use the expressways, to avoid being weighed. One reason (among others) for the relatively low load factor among TI's is that the national highways are too narrow or winding for them, so they cannot so easily avoid the weigh-stations.
13. Using all the above, it is estimated that the trip matrix costs operators about 1,184 billion won (US$I.5 billion). It will be recalled that this excludes all intra-urban trips and inter-urban movements ofbulk and perishable goods and other commodities unsuited to consolidation. While accounting for about 71 % of the national total ton-krns, it accounts for 36% of the total of truck operating costs, since the excluded short urban hauls in small vehicles are by far the most costly perton-km.
14. Departures from the base case of the model are described below in regard to each hypothesis.
54
Higher tractor-trailer productivity
15. Two changes are introduced: TT waiting/loading/unloading time is reduced from 4 hours to 3, and the TT load factor on loaded trips is raised from 70% to 75%. In consequence the average distance driven per year by TTs rises to 96,000 km, the number needed drops by 187, and TT operating costs are reduced by 18 billion won per year.
Shippers ship less frequently, switch to TTs
16. In response to the 10-15% cut in TT rates, freight now moving in small trucks is consolidated by the shipper. The saving is estimated with respect to present private and commercial non-specialized trucks (i.e. excluding dump trucks and tankers) of more than I-ton capacity, whose total supply cost in the base year (from Table Al(c) column 39) is 1,700 billion won. Thus:
S 0.5 X
(For an explanation of the equation's form, see para. 27 below.)
Door-to-door consolidation
17. First, the break-even distance is determined, comparing the VOC for an 8- or II-ton truck with that of a IT driving an extra distance of 30 km for secondary pick-ups and deliveries. It is found to be 50 km. The total saving, if 50% oftonnage in the eligible distance/truck-size cells switches to TTs, is 43 billion won. The number ofTTs needed jumps from 4,500 to 7,400. Alternatively, if the extra distance for pick-up and delivery is only 20 km or less, consolidation in TTs will be the cheaper option over all distances; the saving would be 47 billion won (again assuming that only halfof all eligible traffic would in fact switch).
Terminal consolidation
18. The terminal study provides the following estimate of the costs of processing freight through a terminal:
(a) Storage of miscellaneous manufactured goods: 440 won per ton or m3 per day.
(b) Loading/unloading: 330-670 won per ton (about 1,000 won ($1.25) per labor hour, 2-3 tons per man-hour). Four such operations are normally required on each trip.
(c) Inventory carrying cost: 2-3% per month interest on value per ton ranging from 0.4 million won ($500) for newsprint to 3 million won($3,750) for red peppers.
55
An average cost per trip, assuming a 24-hour wait at each terminal, would be:
storage 800 loading and reloading 1,300 - 2,700 inventory 1,000 - 3,000 Total 3,100 - 5,700
19. The pick-up and delivery to and from the terminal will be done by 2.5-ton and 4.5-ton trucks, at the cost which corresponds to an annual mileage of 20,000 km., i.e. 320 and 195 won per ton-km respectively (40¢ and 24¢).
20. An important consideration is the value that shippers, by their present choice of truck size, show they attach to the smallest sizes. It is fair to deduce that this value is at least equal to the difference in VOC between a 2.5-ton truck -and an 8- or II-ton truck (the largest readily available at present). For this comparison we hold haul distance constant and assume that the empty running rate would be the same at that distance for any truck size. (As for load factor, we assume it does vary with truck size, though we cannot relate it to haul distance.) It is rational to prefer the smallest truck only if the identifiable costs ofputting together larger shipments added to the perhaps intangible convenience premium that the shipper sees in the small truck, would exceed the VOC saving offered by a larger truck. The cost and inconvenience ofputting together larger shipments (mostly storage and inventory carrying cost) grow with the distance, as the probability of a shipper having sufficient sales volume to fill larger trucks declines. For a 30-km trip, the VOC difference between a 4.5-ton truck and the average of an 8- and II-ton truck is about 2,000 won per ton, equivalent to about one day's storage and inventory carrying for middle-value commodities. Over long distances such as Seoul-Pusan (420 km,) the difference is 6,000 won per ton, equivalent to 3-4 day's inventory build-up. For a shipper who now uses 2.5 ton trucks rather than 4.5-tonners, the inventory build-up cost is likely to be at least twice as great.
2 1. The residual difference in the VOC costs between 2.5 tonners and 8-11 tonners can then be interpreted as a lower limit to the elusive convenience premium. It comes to 4,000-4,500 won ($5-6) per 2.5 ton truckload. Unfortunately we have no basis for estimating its upper limit. A consistency check on this interpretation of the cost differences is that by implication a shipper now using 2.5-tonners in the Seoul-Pusan market would take at least nine days to build up sufficient inventory to fill a tractor-trailer; this is plausible.
22. When a shipper is given the option of an LTL service through a terminal, he no longer needs to think about building up inventory. Only the qUality/convenience premium remains to add to the straight cost comparison. Using the lower limit given above, the following tests of break-even haul distances were made, as a function ofpick-up and delivery (PU/D) distance and terminal costs.
PU/D Break-even distance Distance Terminal Total Each End Costs 281 S.01 savrng
(km) (won per ton) Shipment Shipment (bn won)
5 3,000 50 100 134 5 5,000 100 200 48
56
10 3,000 150 200 39 10 5,000 175 300 22 20 3,000 300 350 7
23. The savings shown above assume that 100% of traffic beyond the break-even distance switches to the L TL terminal service. In reality a small percentage will undoubtedly not switch because the origin or destination is too far from a terminal or because of quality features our model does not capture. While some shippers will no doubt value the convenience of small trucks more than the lower limit, the new services will offer better quality, since the shipper will no longer have to wait to fill a truck. These considerations will offset one another.
24. The third of the above model runs is attached (Table A3) as an example.
Operating area restriction
25. The estimate of the theoretical maximum saving given in para. 6.32 refers to the area fleet, which now (1983) numbers 38,700 trucks (Table Al(a)). Their fixed costs, using the values given in Table AI(b), come to 354 billion won per year. The 21-28 billion won saving cited is 6-8% of this amount.
Own-account trucks operating for hire
26. The benefit estimate given in para. 6.42 is obtained by equating the annual mileage of private trucks in Table Al(a) to that of commercial trucks. The categories in question are the general-purpose trucks over I-ton except tractor-trailers, which already achieve comparable, or actually superior, annual utilization. Ton-km performed are held constant, as well as load factors and empty running rates, in calculating the number of trucks needed. It will be noted that by far the biggest difference in annual mileage between private and commercial trucks is found with 4.5-tonners: 32,100 km against 70,5000 km. It is in improving the use made of these middle-size trucks that the greatest savings can be made.
------------
- - - - - - - - - - - - - - - - - - - - ---- - -- - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - --- --
------------ ---------------- -- - - - - - - - - - - - - - - - - - - - - - - --
------------
57
Table A3 TERMINAL CONSOLIDATION (WITH IMPROVED T-T PRODUCTIVITY)
TRUCK CAPACITY
(TONS)
2.5 4.5 8-11 TT
2.5 4.5 8-11 TT
TOTAL
AVHAUL (KMS)
2.5 4.5 8-11 TT
HRS USE
AVGSPD HRSffRP EMPTY % CYCUWK EXC. TT KMSNR EXC.TT
2.5 4.5
8-11 TT
2.5 4.5 8-11 TT
2.5 4.5 8-11 TT TERMINAL GENCST
TOTAL
DISTANCE RANGE (KMS) ____ 9-.?9 ____ :1.99 ____ :I~Q ____-}W ___ 3~0___ :~0Sl___ :3.?9 ___ :49Q ___~~O______________ I9!~\L_ 2.5T LOADS> 150 KM 4.5T LOADS >200 KM GO VIA TERMINAL & T - T
TONS (f000) 12860 6823 3405 o 0 o 0 0 0 23,088 4758 5244 2943 1450 0 o 0 0 0 14,395 7719 9496 5615 5815 5553 3403 3773 6261 781 48,416 5868 6935 2507 4364 6033 2473 3887 3777 375 36,219
SHARE OF DISTANCE MARKET (%) 41 24 24 0 o o o o o 19 15 18 20 12 o o o o 0 12 25 33 39 50 48 58 49 62 68 40 19 24 17 38 52 42 51 38 32 30
100 100 100 100 100 100 100 100 100 100
TON - KMS (MN) 30 70 120 170 220 270 320 370 420
1,272 386 478 409 o o o o o o 1,109 143 367 353 247 o o o o o 8,551 232 665 674 989 1222 919 1207 2317 328 6,498
30_1__ 742__1327__668 -----.11.44 1397 ~
17,431 936 1995 1736 1977 2549 1587 2451 3714 486
~~
PERFORMANCE MODEL PER WK 72 AVAILABILITY 0.80 EXC. TT 0.90
W AIT+LOAD+UNLOAD HRSffRIP 6 EXC. TT 3
22 41 47 49 50 51 52 52 53 1.38 1.72 2.57 3.46 4.36 5.26 6.17 7.07 7.98
45 37 35 33 33 33 32 32 32 8 7 6 5 4 4 3 3 3
12 \0 8 7 5 5 4 4 3 20 41 60 71 73 90 80 92 05 34 66 90 III \03 126 120 139 18
V.o.c. (WON) AVGKMS TRUCKMIX% VARfKM FIXIYR (000) (000)
95 7945 33 41 119 8335 42 15 196 10486 68 34 266 13895 , \02 \0------------------------- -- - -- - - - _..
COST PER TON - KM (WON) 320 164 124 111 108 97 \03 95 90 195 \03 79 71 69 63 66 62 8 128 70 55 51 50 45 47 45 3
72 44 37 34 35 32 33 31 3-----------------------------------------------------------------------.----_.COST PER YR (BN WON)
123 78 51 o o o o o o 253 28 38 28 18 o o o o o III 30 47 37 50 60 42 57 \03 14 441 13 21 II 25 46 22 41 44 5 228
24 28 11 12 9 84 4 3 1 0 \0
194 184 127 121 138 75 III 157 20 1,127
TOTAL COSTS, BASE CASE 1,184 __________________ ~~Y~Q§ ~t:! ~Qt:! !'?!3-_'(E_~)___________________________________ ~7_ .
59
Table A3, page 2 TERMINAL CONSOLIDATION (continued)
TRUCK CAPACITY
LOAD 2.80 5.00
10.20 17.25
TOTAL
2.5 4.5
8-11 IT
2.5 4.5
8-11 IT
TOTAL
0-50 -100 -150
VEH-KMS (MN) 251 271 223
52 117 108 41 103 101 19 45 27
362 535 458
ONS PER TRUCK PER DAY 3.52 3.53 3.14 6.29 6.30 5.16
12.82 12.85 11.45 32.53 31.05 25.83
NO. OF TRUCKS NEEDED
12512 6623 3709 2592 2851 1795 2062 2530 1679
549 680 2%
17715 12684 7479
CONSOLIDATION H/E DISTANCE:
DISTANCE RANGE (lans) -200 -250 -300 -350
0 0 0 0 74 0 0 0
146 179 134 175 65 115 57 107
284 293 191 281
(AVG LOAD X LOADED TRIPS PR DAY) 2.66 2.15 2.16 1.62 4.75 3.83 3.85 2.90 9.69 7.82 7.86 5.92
22.95 16.53 16.62 13.34
(TONS/365ffONS PER TRUCK PER
0 1045 2054
579
3678
DAY/AVAIL %) 0 0 0 0
2432 1482 1111 453
34542 1935
0 0
2183 887
3070
-400
0 0
335 119
454
1.63 2.91 5.93
13.38
0 0
3614 859
4473
COSTS FOR ONE T-T TRUCKLOAD (000 WON)
-450
0 0
47 13
61
1.63 2.91 5.95
10.05
0 0
450 114
563
PICKUPIDELIVERY DISTANCE EACH END (KM) HAND LNG, STORAGE & INVENTORY CARRYING COST:
2.5TR 166 198 257 326 411 4.5TR 101 124 163 209 263
8-11 TR 66 85 115 149 188 IT 37 53 77 100 131
VIA TERMINAL TT+2.5 199 215 239 262 294 TT+4.5 156 172 196 219 251
10 3000
452 292 211 151
566 364 261 181
608 394 285 200
650 423 309 237
313 270
343 300
362 319
399 356
IMPUTED GENERALIZED COST (2.5T VOC to 8T:VOC - 4000 WONffON INVENTORY COST)
2.5- 8T 30 30 4.5- 8T 0 0 SAVING (INCL GEN COST) S-G,2.5 -64 -47 S-G,4.5 -55 -48
(4.ST VOC to 8T:VOC - 2000 WONffON INVENTORY COST) 30 30 30 30 30 30
0 0 0 0 0 0 30
0
-12 -33
-34 -10
-87 -13
110 23
194 65
216 75
220 67
TOTAL
744 350
1,260 566
22,844 8,283
18,486 5,527
55,140
60
Quantity Restrictions on Y ongdals
27. The welfare loss to customers of the yongdals' extra margins (which is the same as the benefits to generated traffic if they were eliminated) can be calculated by the formula for the Marshallian triangle:
where: L welfare loss
L E. dP . Q. 0.5dp P
E price elasticity of demand
dr proportional price change P
Q = quantity (i.e. ton-km) supplied before the price change.
The left-hand portion (E . dP . Q) is the quantity of suppressed traffic and the right-hand portion P
(0.5dP) is the average loss per ton-km. The equation can of course be simplified to O.5E(dP)2. PQ, PQ being the present revenues.
P Thus the monetary value of the welfare loss will be (in won):
x 147 x 0.14J xOA4
[
(elasticity) (margin/revenue (total ratio) revenues)
28. The same equation applies. The total fees are about 40 billion won on total revenues of some 760 billion won, i.e. the margin is about 6%. The welfare loss is therefore (in won):
2 9 0.5 x [0.3] x 0.06 x 760 X 10 {OAt] xt09
0.5 lO.68
61
Appendix 3
ECONOMIES OF SCALE TO TRUCK FLEETS
Background
1. Regulators in many countries have justified entry restrictions and price controls on the grounds that larger truck fleets are more efficient, so would drive out smaller fleets if free to do so. The following analysis is included here to show that the Korean evidence, such as it is, does not bear this out.
Hypothesis
1. Larger truck fleets offer economies of scale, i.e. lower cost per km for comparable vehicles.
Evidence
2. Cost data from the company survey were regressed against both the numbers of company-owned trucks and the total number of trucks, company-owned and owner-operators (ii-ips), in each fleet. The companies did not provide individual costs per truck or even per size category, so to arrive at comparable costs, the aggregate expenditure per company (the raw data) was allocated among the trucks of different sizes in proportion to their numbers and the consultants' synthesized estimates of the average cost per vehicle-km for each of four size categories. Several firms had to be eliminated from the sample when this procedure showed their costs per veh-km to be unrealistically high or low (above US$1.50 or below 10¢). The remaining sample consisted of 30 observations, each being the average cost per vehicle-km for company-owned trucks in a given size category (2.5, 4.5, 8 or 11 tons capacity), spread among 16 companies. The effect oftruck size on cost was eliminated by dividing each observation by the mean cost for its size category.
3. The regression ofthe resulting cost ratios against the size of fleet (Table A4 and Figure 3) showed no correlation whatsoever. With respect to the number of company-owned trucks per fleet the R2 was 0.03 (not statistically significant), and with respect to the total fleet size the R2 was 0.00.
Conclusion
4. While the sample size was small, the complete absence of any correlation is cogent evidence that there are no economies of scale to fleet size among Korean commercial truckers. Most studies of fleets in other countries have reached the same conclusion. 61
Spady, R.H., and Friedlaender, A. F, "Hedonic Cost Functions for the Regulated Trucking Industry," The Bell Journal (Spring 1978).
61
--2.00
1.80
1.60
1.40
1.20
1.00
0.80
0.60
0.40
0.20
o
2.00
.... ~ Ot:: 1.80r.nr.n ;..::f-U~ ~f- 1.60 f- ....
00Zr.n 0;":: 1.40 :::E U <~ ~::; 1.20 .,.-l
::t:< ~.:r;: >0 1.00.:r;: .... ~Z.::0..< f-~
0.80~:::E Uo Of-Z . )::>- 0.60
~~ ~:::E
°u 0.40.::0..0
cj.:r;:
~~ 0.20
.... ~ 0 .... 0Or.n (::~;2:J
62 Figure 3
OPERATING COSTS OF COMPANY -OWNED TRUCKS IN RELATION TO COMPANY FLEET SIZE
-f
-.
1: i i
I I
~i--
~f-
o
x x
v
X X
X
X x tR? = 0.01..
x x --r--)j -x '" xx x x X
x x x
x x X
10 20 30 40
x
-f
-l
-l
x -f- x
-f
x
--".
X X
X
X
X X R2=0. C05
x X x x ~ x
rx,xx
x x
X X
x x
x
I I I I I
o 20 40 60 80 100
63
64
Table A4 OPERATING COSTS OF COMPANY-OWNED TRUCKS
RELATIVE TO TRUCK SIZE AND FLEET SIZE
Company
4-1 4-4 5-1 5-2 6-2 6-6 7-1 7-3 7-9 7-14 8-5 8-9 8-13 12-7 12-9 12-10
Company owned Fleet
10 5 12 6
35 7 9 30 7 3 5 9 9 5 3 15
No~ 16 Mean 1
Total Fleet
106 5
73 52 35 35 83 46 34 6 53 40 61 40 17 105
16 331
227
203
179 280 283 333
6 369
335
190
138 675 190 194
315 447
296
281 301 533
345
175 678
285 252
469
10 1
5
237 240 440 418
271
9
0.90
0.81
0.71 1.12 1.13 1.33
V.O.c. Divided by Size Mean 4.5T 8T
1.01
0.820.57 0.74 1.45 0.75
1.37 1.60 0.94 1.30 0.42 2.04 0.57 0.59
0.95 1.35
0.84 0.89
65
Appendix 4
INFORMATION SYSTEM FOR MONITORING AND EVALUATING THE PERFORMANCE OF THE TRUCKING TNTHISTRY
A. Objective
1. The purpose of the proposed infonnation system is to provide the government with data it needs to monitor the perfonnance of the trucking industry, with regard to the profitability of operations for the trucker and the cost and quality of service to the user. In this way it will be better able to judge when and where government intervention, if any, is called for in order to promote national objectives, and to decide what fonn such intervention should take. In particular the infonnation system should monitor the need for action to over-ride market forces as to the number and capacity of trucks, the level and structure ofrates, and the scope of truckers' operations,
B. Necessary Indicators
2. The Ministry of Transport (MOT) should calculate average values of the following indicators at regular intervals. In some cases the basic data can be collected directly at the national level, while in others the basic data will have to be compiled at the provincia1level and then aggregated or averaged at the national level. Some items change only slowly, others quite rapidly; the latter should therefore be up-dated more frequently.
For Monitoring Freight Rates
(a) Unit costs: up-to-date data on the main components of vehicle operating cost, for each of the main truck sizes/types
Frequency - fuel national 6 months
tires spare parts (expenditures per year per vehicle)
- interest rate - insurance
wages of drivers wages of workshop labor
(b) Vehicle productivity indicators: the link between expenses incurred periodically and the cost per ton-km carried
ton-km per month derived from either: province 3 months • total kms driven • empty mileage • payload capacity
66
• average load factor on loaded trips or
• average number of trips per month • average haul distance • average load per trip
directional imbalances of freight flows on main routes • traffic in the return or secondary
(i.e. less heavily trafficked) direction as % of traffic in the principal (i.e. more heavily trafficked) direction
seasonal variation • ton-km per truck per quarter
as % of the annual total
For Monitoring Truck SuppLy
province 12 months
province 3 months
A sharp increase in any of the following indicators would suggest an excess of demand relative to supply:
(a) Volume indicators
number of applications for new operating licenses growth rate of total fleet ton-km share of own-account trucking in total market
vehicle utilization rates (load factors, etc.)
(b) Price indicators
prices of used trucks freight rates actually paid on non-contract business
(c) Specific physical measures
delays between time when user requests truck and vehicle arrives to load build-up of inventories in warehouses average time to unload ships or trains
Provincial Offices MOT
"
Sample of users Sample of dealers
Sample of users
"
3. The above indicators, none ofwhich is conclusive evidence on its own ofunsatisfied demand, when taken together would provide independent corroboration (or would refute) the arguments put forward by license applicants, the opinions ofthe truckers associations, and
67
complaints received directly from users. In the above list, the indicators are shown in order of increasing cost to compile. On the other hand the least costly indicators are also the least specific and possibly therefore the least conclusive as evidence.
4. Level and frequency of compilation: The productivity and price indicators would be compiled at the provincial level regularly every 3 months. The specific physical measures would be compiled "ad hoc", as required by the number, type and geographical concentration of applications received.
C. Organizational Arrangements
Own-account Trucking's Market Share
5. The share of own-account (private) operations in the national market for road freight transport can be estimated at relatively low cost if the operators of own-account fleets are required to report the ton-km they have performed, each time they renew their vehicle's operating license. The provincial and county offices who process these renewals would send in this data to the MOT as part of their regular reporting.
Commercial Trucking
6. As regards commercial trucking, the flow of information required can be summarized as follows:
companies and owner-operators
(truckers associations)
MOT Land Transport Bureau
Companies
7. A sample of 1 in 10 trucking companies and owner-operators would be selected randomly once a year by each provincial/city office, so that the same respondents would report for each of the four quarters of a given calendar year. The sample would be stratified to distinguish holders of "area", "special", "route" and "yongdal" licenses.
68
8. Every 3 months the sample of respondents would provide the provinciaVcity government, through the truckers associations if that is deemed convenient, the following:
(a) Numbers of trucks of each type ("yongdal"; general-purpose 2.5-ton, 4.5-ton, 8-ton, II-ton and tractor-semitrailer combination; dump truck, tank truck; and others)
(b) Data on output for each subgroup and vehicle type, as in the accompanying schematic (note that "capacity offered" means "Total kms driven" x "Payload capacity").
69
Raw Data - Quarterly Totals for Sample Fleet
ChejuI
...I
TaeguI
PusanI Seoul
Area 2.5 .. TT
1. No oftrucks 2. Kms loaded 3. Kmsempty 4. Total kms 5. Capacity (TK) -offered 6. TK performed
in principal direction 7. TK performed
! in return direction 8. TK performed
'--
-
-
1. No. of trucks 2. Kms loaded 3. Kmsempty 4. Total Kms 5. Capacity (TK)
offered 6. TK performed
in principal directiOl 7. TK performed
in return direction
70
9. The data should be sent to the province/city government within one month of the end of each quarter.
Office of the Province/Special City
10. Every 3 months the province/city would aggregate the above data and calculate estimated total or average values for the fleet resident in its territory, expanding from the sample to the total fleet.
11. It would then send the aggregate data to the MOT, within a month and a halfof the end of each quarter. The fonnat for the tables is suggested in the accompanying schematic.
Cheju
I
PusanI SeoulI
Total
Area Truck size Total
1. No. of trucks in fleet 2. No. of trucks sampled 3. Sample size (2 as % of 1)
Fleet totals 4. Total kms driven 5. Emptykms 6. Capacity (TK) offered 7. TK perfonned in principal direction 8. TK perfonned in return direction 9. TK perfonned in total
10. Directional balance (8 as % of7) Average per truck
11. Total kms driven -
12. Empty kms 13. Empty as % of total (12 as % of 11) 14 r;m;:l~itv (TKI offf':Tf~rl
c--
... and the same for special, route, and yongdal trucks
71
Once each year the provincial offices would submit, additionally, a table summarizing the seasonal variations for each category of truck.
Land Transport Bureau of MOT.
12. The Land Transport Bureau would aggregate the data into national totals and calculate average values of efficiency indicators and trends over time, by 2-112 months after the end of each quarter. The basic format for the tables would be the same as for the provincial aggregates.
13. Every 6 months the Bureau would also collect price data on the components of vehicle operating cost, including the prices of used trucks.
14. It would distribute the results to the Transport Co-ordination Bureau for information.
Physical Measures of Excess Demand
15. Periodic campaigns would be conducted jointly by the provincial offices and the truck associations under instructions from the MOT, on specific routes, in specific cities and ports, or for specific commodities.
16. Once the information system is operating routinely, it would be implemented entirely by the regular staff of MOT, the provincial governments and the truckers associations. However, during the start-up phase training seminars could be organized for the staff designated to assume responsibility for the system, to clarifY the system's objectives and to familiarize the staff with the potential costs and benefits of collecting more or less data. Another task would be the designing of the forms to be filled out by the regular suppliers of data and the drafting of an explanatory manual.
72
BmLIOGRAPHY
Baum, H., "Possibilities and Limits of Regulation in Transport Policy." 1983
Beesley, M.E., "Assessing Transport Regulation in Developing Countries," World Bank, 1984 (TRP internal report)
Boyer, K., and Shepherd, W. (ed). "Economic Regulation: Essays in Honor of James R. Nelson." MSU Public Utilities Papers, 1981.
Foster, C. (Chairman). "Road Operators' Licensing," Report of the Independent Committee of Inquiry. HMSO, 1978.
Friedlaender, A. "The Dilemma of Freight Regulation." 1969.
GEIPOT, "Avaliacao da Regulamentacao Economica do Transporte Rodoviario de Carga," Brasilia, 1980.
Phillips, A., (ed.) "Promoting Competition in Regulated Markets. Studies in the Regulation of Economic Activities." Washington, D.C., 1975.
Spady, R.H., and Friedlanender, A.F. "Hedonic Cost Functions for the Regulated Trucking Industry." The Bell Journal (Spring 1978).
World Bank and Netherlands Institute of Transport. "The Determinants ofModal Choice in the Freight Transport Market." 1984.
73
Tatiana Daza C:lKoreatrk2.doc 3/28/005:26 PM
Related Documents
![TRUCKING INDUSTRY · [ 6 ] TRUCKING INDUSTRY ENVIRONMENTAL BEST PRACTICE GUIDE ThE AUSTRALIAN TRUCKING ASSOCIATION The Australian Trucking Association (ATA) is the national peak body](https://static.cupdf.com/doc/110x72/5e6e335355a7c4346213f0b4/trucking-6-trucking-industry-environmental-best-practice-guide-the-australian.jpg)
