the freight transport sector in the 1990s

This article was downloaded by: [York University Libraries]On: 12 November 2014, At: 07:19Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number:1072954 Registered office: Mortimer House, 37-41 Mortimer Street,London W1T 3JH, UK

South African Journal ofEconomic HistoryPublication details, including instructions forauthors and subscription information:http://www.tandfonline.com/loi/rehd19

The freight transport sectorin the 1990sTrevor Jones aa University of NatalPublished online: 04 Mar 2010.

To cite this article: Trevor Jones (2003) The freight transport sector in the1990s, South African Journal of Economic History, 18:1-2, 213-237, DOI:10.1080/10113430309511160

To link to this article: http://dx.doi.org/10.1080/10113430309511160

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of allthe information (the “Content”) contained in the publications on ourplatform. However, Taylor & Francis, our agents, and our licensorsmake no representations or warranties whatsoever as to the accuracy,completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views ofthe authors, and are not the views of or endorsed by Taylor & Francis.The accuracy of the Content should not be relied upon and should beindependently verified with primary sources of information. Taylor andFrancis shall not be liable for any losses, actions, claims, proceedings,demands, costs, expenses, damages, and other liabilities whatsoever

http://www.tandfonline.com/loi/rehd19

http://www.tandfonline.com/action/showCitFormats?doi=10.1080/10113430309511160

http://dx.doi.org/10.1080/10113430309511160

or howsoever caused arising directly or indirectly in connection with, inrelation to or arising out of the use of the Content.

This article may be used for research, teaching, and private studypurposes. Any substantial or systematic reproduction, redistribution,reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of accessand use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

http://www.tandfonline.com/page/terms-and-conditions

http://www.tandfonline.com/page/terms-and-conditions

213

The freight transport sector in the 1990s

by

TREVOR JONESUniversity of Natal

1. INTRODUCTION

In 1994, South Africa's new democracy inherited a transport sectorthat had undergone major changes over the previous quarter century.In the decade of the 1970s, the freight transport sector had moved irre-versibly into the era of high-mass transportation, most notably of bulkcommodities moving through the newly-constructed ports andlandside infrastructure associated with Richards Bay and Saldanha.General cargo handling by rail, road and sea was overtaken by thecontainer revolution and these flows passed, for the most part, throughfacilities that were up the task, although the domestic transport marketretained significant imperfections that limited the participation of pri-vate commercial road hauliers in long-haul intercity freight move-ments. The decade of the 1980s witnessed a deepening and consolida-tion of these facilities and activities, as well as the final dismantling ofthe road permit system that had so constrained the market penetrationof road transport.

Consequently, by 1990 the South African transport sector possessedmany apparent strengths. World-class bulk ports and rail systemstransferred substantial volumes of low-value primary exports to worldmarkets at low cost; fairly static volumes of general cargo flowedthrough a physically extensive rail infrastructure and increasingly byroad between domestic centres and to and from ports that were ade-quately equipped; world-class national roads linked cities, some of

Volume 18: September 2003

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

214

whose residents enjoyed high levels of private car ownership; andcommercial agriculture was reasonably well integrated into the broad-er transport infrastructure.

Other segments of the economy were significantly less well-served.Many rural communities were excluded from the transport network,some Black commuters were forced to shuttle daily over long distancesbetween dormitory suburbs and urban industries, and other peri-urbanresidents distant from commuter rail services were left to the merciesof chaotically-organised taxi and bus services. This split personality onthe part of the transport sector may not have been an intractable prob-lem for a political economy that sought to serve the interests only of aminority of the overall population, but in many respects the sector waspoorly prepared for the needs of a more inclusive structure, and a moreoutward-looking trading economy, to a point where the Department ofTransport's influential Moving South Africa study identifies "a trans-port system and infrastructure, in fundamental misalignment to thenew national objectives".1

This article seeks to explore certain dimensions of that misalignment.Areas of technical and policy interest emerge in the spheres of bothfreight and passenger transport, but this coverage will be limited tofreight movements effected by the three principal surface modes ofsea, rail and road transport.

2. PORTS AND MARITIME TRANSPORT

After substantial capital-widening investment in the 1970s and consol-idation in the 1980s, the South African ports sector entered the 1990swith assets that were massive by the standards of emerging economieselsewhere in the Southern Hemisphere, but also with a configurationthat was heavily skewed towards bulk-related seaborne commerce.

1. Department of Transport, Moving South Africa (Draft for discussion), Pretoria,1998, p 5

South African Journal of Economic History

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

215

Thereafter port expansion slowed significantly; even a growing real-isation that world-beating bulk ports and mediocre general cargo facil-ities were seriously at odds with the development orientation of a"new" South African economy that was attempting to re-enter themainstream of an increasingly globalised international trading commu-nity, largely failed to translate into new marine facilities or cargo-han-dling systems.2 The limited investment that took place between 1990and 2000 largely served to increase the degree of asymmetry in theports: some extensions to bulk and neo-bulk cargoes3 facilities inRichards Bay were completed, and some downstream steel activitieswere incorporated into the port infrastructure in Saldanha Bay. Bycontrast, no significant new marine infrastructure for general cargo ac-tivity was created, despite tortuous attempts to identify sites for addi-tional cargo-handling capacity in Durban,4 while long-term plans for anew industrial bulk port and potential container "hub" at a virgin sitenear Coega in the Algoa Bay area were greeted with scepticism bymany sections of the commercial fraternity, and would in any eventonly come to consummation well into the first decade of the new cen-tury, as would more ambitious plans to remodel Durban's general car-go infrastructure. Improvements in the general cargo-handling arenawere consequently limited to extensions to landside back-of-port facil-ities, principally associated with the Durban Container Terminal, andmore substantial capital deepening of existing facilities, via investmentin further cargo-handling superstructure.

2. Ibid; Derek Lawrance, Ports and transport logistics in Southern Africa:performance and prospects, Maritime Africa 2000 Conference Proceedings,Durban, 2000

3. Seaborne traffic is most broadly disaggregated into homogeneous bulk cargo,such as coal, grain or sugar, where individual units of cargo cannot be identifiedon the basis of individual mark or number; and heterogeneous general cargo, withindividual mark or number. Neo-bulks are cargoes that have some characteristicsof bulk cargo, but are partially differentiable. Examples would include steel rolls,coils and I-beams, reels of paper or blocks of granite.

4. IEM Study Group, "Port of Durban container handling: Integrated EnvironmentalManagement (IEM) study", unpublished report, Durban, 1996


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

216

TABLE 1

Cargo handled at South African ports for selected years1990-2000

{million metric tons)*

Year

1990199119931994199719981999

RichardsBay52,656,760,669,081,486,486,4

Durban

38,439,646,641,452,752,451,8

EastLondon

2,51,61,63,21,81,61,3

PortElizabeth

5,34,74,74,96,76,76,7

CapeTown

7,17,07,39,89,88,58,3

Saldanha

25,019,724,423,324,926,624,6

Total"traff

130,9129,3150,7151,6177,3182,2179,1

2000 91,8 55,0 1,5 8,1 10,1 26,4 192,9

* All cargoes, including petroleum products. Estimates of oil and petroleummovements from 1990 to 1994 are based on Charlier's work and may beconservative. 1997-2000 data in respect of these products are more accu-rate.

** Total traffic excludes the activities of Mossel Bay, where volumes weretiny before the Mossgas project came on stream from 1992. By the late1990s, about 1,7 million tons of traffic, almost all of it petroleum prod-ucts, was handled annually.

Source: Jacques Charlier, he systeme portuaire sud-africain a Vaube du21eme siecle, Brussels, 1996; Port of Durban Statistics, 1997-2000; Keren Giladi, "The South African oil industry and its rela-tionship with the ports", MBA dissertation, University of Natal,2003 (Oil & petroleum product volumes 1997-2000)

This evolution from expansion through consolidation to slow-down ismirrored by the traffic performance of the South African ports. Afterdoubling from 40 million tons in 1969/70 to 80 million tons by1977/78, total cargo handled by the major ports in the Saldanha-Rich-ards Bay range increased by a further 70 per cent to reach 136 milliontons by 1989. In the 1990s, as shown by Table 1, traffic growth has


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

217

been more muted, with overall volumes rising a little over 40 per centfrom their 1989 level, to stand at some 193 million tons by 2000.

What growth has been recorded over the decade can be attributed prin-cipally to additional port activity in Richards Bay, through increasedcoal exports and further diversification into various mineral productsand neo-bulk cargoes such as steel and forest products. These havedriven the annual traffic base of the port up from 52,6 million tons in1990 to over 90 million by 2000, a level unmatched by any compara-ble port in the Southern Hemisphere. The second source of nationaltraffic growth is less easy to pin down, but emanates largely from theport of Durban, where increased containerised seatrade and expandedcapacity in the local oil-refining industries have raised port activitysomewhat, although annual volumes have fluctuated in a relativelysmall band above 50 million tons from 1997 to 2000. In the remainingports, no monotonic growth trend is discernible in the 1990s.

Despite its slowing growth over the last 10 years, overall South Afri-can port activity remains enormous by regional standards. This is re-vealed by Table 2, which sets out the performance of a representativeselection of leading African and southern hemisphere ports on the bas-es of total traffic handled and container traffic. A consideration of thelatter is vital, as different types of traffic have obviously different eco-nomic impacts.5 In terms of economic linkages and value-added, han-dling a ton of crude oil is not the same as handling a ton of refinedsugar, while both of these differ markedly from the handling of a tonof containerised machinery parts. It is the broad category of generalcargoes, now almost universally moved in standard freight containers,that confer the richest economic linkages on the local economies of theports where they are worked, and on the wider national economieswhere they originate or are absorbed.6

5. Jacques Charlier, Le systéme portuaire sud-africain á Vaube du 21 erne siécle,Brussels, 1996

6. Containerised import-export traffic per capita is used as one of many indicators ofeconomic development. On this basis, South Africa still lags behind theperformance of other comparable middle-level developing economies, suggestingthat container traffic growth may well continue to outpace overall GDP growth.


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

218

TABLE 2

African and southern hemisphere port traffic in 2000*

Port

Richards BayNewcastleDurbanSantosSydneyMelbourneCasablancaAbidjanAucklandCape TownLagosMombasaBuenos AiresDakarPort Louis

Total porttraffic

(m tons)

91,873,955,043,124,622,319,814,613,311,89,18,97,87,24,7

Rank

12J

456789101112131415

Containertraffic

(000 TEUs)

59

129194599913223114345613951782219716149161

Ran

151424J

19768111051312

* Most prominent multi-purpose ports are included. What are not,are one-commodity "industrial" ports, such as pure oil terminals,and single-commodity bulk export terminals.

** 1997 magnitude

Source: ISL, Shipping statistics yearbook- 2002, Bremen, 2001

In terms of indicators of both total port traffic and container traffic, theSouth African ports emerge as colossal by African standards and asdominant within the broad trading community of the Southern Hemi-sphere. Richards Bay tops the league table in terms of total port traffic(followed by the rather similar coal-oriented Australian port of New-


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

219

castle), while Durban occupies a clear third spot on the basis of totaltraffic and is a close second to Melbourne in respect of container activ-ity, but well ahead of Sydney and Santos. Durban does, however, com-mand a substantially more diverse traffic base than that of either of themajor Australian general cargo ports. It is consequently difficult toavoid the conclusion that South Africa (and indeed the province ofKwaZulu-Natal) possesses both the leading multi-purpose general car-go port and the most active diversified bulk port on the continent ofAfrica and in the Southern Hemisphere.

The relative heavyweight status of this country's ports should come asno surprise, since South Africa has become a major sea-trading nationin the late-20th Century. The 193 million tons of traffic handled in2000 represents approximately 3,6 per cent of world seatrade in ton-nage terms. In terms of "real" sea transport activity, traffic passingthrough local ports generates some 12 300 million ton-miles of mari-time freight activity, or over six per cent of global activity - a perfor-mance that places South Africa within the top twelve nations on the in-ternational maritime-trading league table. This country's share ofglobal maritime activity consequently exceeds its share of global GDPby more than twenty to one

All of this appears to signify a vibrant, growing and sophisticated portssector appropriate for the needs of a middle-income developing econo-my. Appearance and reality may well not coincide in this regard; thereis growing evidence that the facilities and services provided by theports are seriously out of line with the development needs of the econ-omy. For the "old" South Africa, growth through penetration of rela-tively anonymous bulk markets made considerable sense. Consequent-ly, the creation of world-class bulk transport facilities also made emi-nent sense. They make a great deal less sense in the context of a "new"economy, whose principal development drivers are the growth of man-ufacturing output and the expansion of manufactured exports, ener-gised by the ability to acquire imports at minimum cost.7 What this de-

7. Department of Transport, op cit


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

220

velopment path requires from the nation's ports are general cargo-han-dling facilities of the requisite quantity and the highest quality, at thelowest price. Sadly, in terms of all three of these desiderata, the per-formance record of the South African ports is well below the world-class benchmarks achieved in the bulk arena.

The transport and handling of general cargoes is now dominated to anoverwhelming degree by containerisation and unitisation, so that theefficiency of general cargo-handling facilities is essentially determinedby the existence and maintenance of suitable container terminals anddistribution systems. When South African maritime transport joinedthe "container revolution" in 1977, port capacity well beyond the traf-fic parameters of the time was provided for. This installed capacity(both in terms of the basic marine infrastructure of the various port ter-minals and the equipment and logistic networks required to move box-es between the terminals and their hinterlands), remained reasonablyadequate until the early 1990s when national traffic levels were ap-proaching the one million TEU8 level. Since then the container-han-dling problems of the South African ports have escalated.

This problem has several dimensions, affecting both the demand andthe supply sides of the transport market mechanism. On the demandside, the re-entry of the economy into the mainstream of the interna-tional trading community following the first democratic elections in1994, precipitated a surge in seaborne container volumes, as the lead-ing liner carriers returned to the South African trades, and as containerpenetration rates rose from levels that had been low by internationalstandards. In the leading container port, Durban, container traffic roseby 23 per cent in 1993-94, by 14 per cent in 1994-95, and by a further10 per cent in 1995-96. These booming volumes threatened to over-

8. Container traffic levels are conventionally measured in terms of twenty-foot-equivalent-units or TEUs. One TEU represents a standard 20 foot (or 6 metre)freight container. Since forty-foot boxes are increasingly supplanting twenty-footcontainers, the number of physical box moves through a port is lower than thenumber of TEUs.


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

221

whelm terminal capacity, and produced the inevitable consequences ofberth delays, cargo distribution hold-ups and vessel queues outside theports. Demand growth slowed somewhat in the late 1990s, but by2000 the South African ports handled a total of 1,80 million TEUs (ofwhich Durban alone accounted for 1,13 million TEUs), with the samebasic container quays that had been constructed in 1977, albeit sup-ported by an enhanced cargo-handling superstructure. These enhance-ments have, however, not raised cargo-handling productivity to ac-ceptable international levels, nor have they alleviated the congestionthat remained a semi-permanent feature of the general cargo-handlingscene to the end of the decade.

At the same time as these demand-supply pressures were building, aseries of exogenous supply-side changes were unfolding in the interna-tional liner shipping industry. When container operations first super-seded conventional breakbulk shipping services, the routes over whichvessels operated changed relatively little; many liner vessels continuedto ply north-south, region-to-region routes. From 1990, this pattern haschanged significantly, with most high-density liner routes now organ-ised around skeins of round-the-world or long-swing pendulum servic-es,9 operated by collaborative consortia of shipping lines, and connect-ing at a limited number of "hub" ports, from which radiating "spoke"services feed cargoes to secondary ports. Two powerful concomitantsof these arrangements are the deployment of larger, deeper-draftedcontainer vessels, and a reduction in the number of port-calls in anydiscrete geographic region. These arrangements first took root on thehighest-density northern hemisphere trade routes, but many southernhemisphere services have followed suit, albeit with smaller vesselsmore suited to lower levels of demand. The consequences for ports aresubstantial, as the container "majors" attempt to concentrate their ac-

9. Many of these are de facto round-the-world services. Container vessels may runfrom the east coast of the United States, across the Atlantic to a single port of callin the north-west Continent, through the Mediterranean and the Suez Canal toSingapore, thence onward to Japan and finally to a single port on the west coast ofNorth America, before reversing direction.


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

222

tivities at a single regional "hub", that acts as both a sink for nationalcargoes and as a distribution centre for transhipment traffic. Successful"hub" ports therefore have to provide capacity beyond the dictates ofnational demand, hence attainment of "hub" port status becomes inor-dinately difficult in a capacity-constrained situation. That is exactlythe situation in which the South African ports in general, and the dom-inant port of Durban in particular, have found themselves since theearly 1990s. Durban is the only South African port with any preten-sions to "hub" status, and it has been favoured as the principal, or insome cases as the sole, South African port of call by many liner opera-tors10 for several years. Preference for Durban has arisen from itsproximity to the economic powerhouse of Gauteng, ample spare ca-pacity on inland rail and road arteries,11 and an excellent network ofport-ancillary industries serving the needs of vessels and theircargoes.12

From the 1990s, these strengths have been threatened by capacity con-straints that have been exacerbated by a reluctance on the part ofshipowners to shift to less-preferred ports, such as Port Elizabeth, de-spite the blandishments of (uneconomic) rail-rate equalisation to bringrail costs on the longer Cape-Gauteng routes in line with the shorterroute from Durban. The overall picture to emerge from this account isof container facilities that have lurched from capacity crisis to capacitycrisis since 1993, with investment spending well below long-term cap-ital requirements.13 The response of the port authorities (who were stillalso the terminal operators at this time) has been muted to a point ofnear-catatonia. In Durban, a reconfiguration of the port to permit con-tainer expansion into a port segment hitherto dedicated to neo-bulk ac-

10. Lloyd's List Africa Weekly, Johannesburg, 199911. On the Durban-Gauteng rail line, the highest utilisation (approximately 50 per

cent) is experienced on the Durban-Pietermaritzburg segment and in the environsof Johannesburg. Elsewhere, utilisation levels of 28-30 per cent were the order ofthe day in the mid-1990s.

12. Trevor Jones, The port of Durban and the Durban metropolitan economy,Economic Research Unit, University of Natal, Durban, 1997

13. Department of Transport, op cit, p 50


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

223

tivities will create a capacity-growth window of five to seven years.However, if container seatrade continues to grow at the annual averageseven per cent rate recorded from 1992 to 2000, major new infrastruc-ture will have to be created, preferably on the eastern seaboard, beforethe end of the first decade of the 21st century.

The final dimension of the South African ports sector to give cause forconcern is the cost of port functions to users. The Moving South Africastudy demonstrated that waterfront charges in this country's ports arehigh by international standards when measured against comparableports in both developed and developing countries.14 This blanket com-parison also masks some major intra-port price-cost skewness. Histori-cally, marine charges (such as port dues and tug charges) have beenlow relative to both foreign harbour tariffs and underlying costs. Thisunder-recovery has been more than made up by cargo charges thathave been grossly in excess of cost.15 The principal culprit in this re-gard is the single item of ad valorem wharfage, which was levied dif-ferentially on imports and exports, and which was, as the name sug-gests, driven by commodity value rather than by considerations ofcost. This economically inefficient pricing mechanism was intended tofinance the cargo-working infrastructure of the ports, and throughoutthe second half of the 20th century it was the principal source of bothharbour revenues and substantial overall port profits. In overall tariffterms, the South African ports have therefore been cheap for ships butexpensive for their cargoes. This unfortunate asymmetry militatesagainst marginal parcels of cargoes, most notably the categories ofhigher-valued general cargoes that top the agenda in the state's manu-facturing-led development strategy. These pricing anomalies are un-likely to survive future port privatisation, or a fuller separation of land-lord and operating port planned by the port authority; but they havedominated the port pricing landscape for the last 50 years.

14. Ibid, p 4615. Trevor Jones, Coastal sea transport and intermodal competition, National

Institute for Transport and Road Research, Pretoria, 1985


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

224

In overall terms, the South African ports sector has developed a dis-tinct schizophrenia: a top-quality bulk infrastructure offers first-worldservices to bulk exporters and importers at competitive prices, whilecontainer and general cargo terminals, operating at their elastic limits,offer a poor service at high cost. Sadly, the opposite strengths andweaknesses would better serve the development needs of the economy.

3. RAIL TRANSPORT

By 1990, rail transport16 was in incipient eclipse. An exuberant phaseof expansion in the 1970s, once again driven by burgeoning bulk ex-port traffic, had been followed by something very close to stagnationduring the 1980s. The 1990s were to witness a decline in the substanceif not necessarily in the full arithmetic of rail performance: total routekilometres of track under the control of Spoornet, the rail division ofTransnet, fell marginally as certain low-density branch lines wereclosed; total freight conveyed fluctuated in a broad band between 168and 186 million tons (below the peak reached in 1981); but below thesurface of this aggregated tonnage, serious inroads into the carriage ofhigher-rated general cargo were made by privately-owned commercialroad carriers that were freed from the shackles of the restrictive roadpermit system. By the end of the decade, rail had developed the samechronic bipolar disorder as the ports, with a bulk export infrastructureof genuine world class co-existing with a general cargo-handling net-work of increasing fragility, and a traffic base heavily skewed towardslow-valued bulk staples. Once again, the overarching development im-peratives of the South African economy would have been better servedby a reversal of these symptoms.

If South Africa's public rail network, rolling stock and output levelsare compared with those of other significant economies in Africa or

16. Coverage of rail transport is limited to the network and activities of Spoornet, therail division of Transnet Ltd. The few privatised lines, as well as rail transporteffected "in house" by various mining and industrial organisations, are ignored.


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

225

TABLE 3

Rail transport size and scale indicators in selected African andsouthern hemisphere countries in the mid-1990s*

Country

Argentina**BrazilChile

Networkroute-kms

340592266482472

South Africa***204413NigeriaEgyptAlgeriaAustralia****

30544810429016492

Staff

670006164522374714011346910651584728765

Loco-motives

9921805132

35472008352312007

FreightWagons

3282352039

347133645no data

3102572

12807

FreighttonsxlO6

No data105,034,53

180,130,1652,411,80

76,10

Freightton-km;

xlO 6

786045664

96795591

no datano datano data40000

* Years of enumeration vary across countries, but are generally drawnfrom the 1990-95 period. The South African staff level is based on 1998data.

** Over 10000 kilometres of Argentina's rail infrastructure comprise nar-row-gauge lines.

*** Excluding private lines operated by various mines and industries; 1998data.

**** The Australian information represents an amalgam of the activities of theAustralian National Railways (ANR) and State operations in South Aus-tralia and New South Wales. Ton-kilometre calculations were furnishedonly by the ANR, but have been aggregated heroically for the wholecountry.

Source: Business Rail Report, International data bases, Railway Gazette In-ternational, Sutton, 1998; RailRoad Association of South Africa,2000 (website www.rra.co.za)


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

226

elsewhere in the southern hemisphere, as shown in Table 3, South Af-rica shows up strongly. This country's basic rail backbone, measuredin route-kilometres, stands well below that of Argentina's rail networkand somewhat below that of Brazil, but exceeds those of other SouthAmerican, African and Australasian economies. In terms of locomo-tives, freight wagons, freight tons carried and freight ton-kilometres of"real" activity, South Africa eclipses all other countries enumerated, inmost cases by a wide margin. By contrast, staff numbers employed bySpoornet are well below railway employee levels in Argentina, Braziland Egypt, despite the generally lower levels of output generated in allthree of these countries. This broadly favourable comparative interna-tional performance of South African rail largely mirrors the domi-nance of the ports sector, and suggests that this country is generallywell equipped, if measured alongside other middle-income developingeconomies, in terms of hard-core transport infrastructure.

The data in Table 3 can also be used to generate various measures ofrail productivity. In terms of freight ton-kilometres per employee,South African rail outperforms Australia by some 70 per cent, Brazilby a factor of three, and others by larger margins. In terms of freightton-kilometres per wagon, this country's rail productivity stands at 54per cent of the level produced by the Australian National Railways (anAustralia-wide calculation cannot be made), and at a little over 80 percent of the Brazilian level. Other useful performance indicators maybe constructed around rail rate levels as measures of the cost of railservices to users. On the basis of 1998 traffic levels, price per ton-kilo-metre weighted across the aggregate South African rail traffic baseemerged as some R0,22. This compares unfavourably with the UnitedStates (R0,ll per ton-kilometre) and Sweden (R0,13), but favourablyalongside many other developed economies such as France (R0,37),the United Kingdom (R0,48) and New Zealand (R0,51).17 Compara-tive data for other middle-level developing economies are not avail-able. In user cost terms, South African rail services are moderately-

17. Business Rail Report, International data bases, Railway Gazette International,Sutton, 1998


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

227

priced by international norms across the full gamut of traffic, while thespecialist Coallink "produces the lowest-cost inland coal transport inthe world at a high level of efficiency".18

FIGURE 1

Evolution of rail freight traffic1976-1999

160

120

e 80-

40

1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998

—•— Total —•— General —A— Coallink —K— Orex

The above account generates a picture of South African rail that issubstantially positive: this country possesses a larger and more pro-ductive rail network that those of many other comparable nations, andoffers its services to users at costs in line with or below those associat-ed with many economies. These broad indicators may, however, con-vey a seriously distorted image of a mode of transport that this paperwill argue is far from healthy, and far from fulfilling the role it shouldperform in the South African economy.

The overall fortunes of rail may perhaps more accurately be represent-ed by the changing composition of its traffic base, rather than by theperformance of nominal or real output. Figure 1, below, depicts theevolving size and shape of total rail freight traffic from 1976, the year



Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

228

when high-mass bulk cargo begun to flow to Richards Bay and later toSaldanha, to 1999. In terms of crude tonnages of freight conveyed,freight activity increased from 141 million tons in 1976 to a peak of188 million tons by 1981, as coal and then iron ore exports made theirpresence felt, but by 1990 total freight transported had fallen back to176 million tons. During the 1990s, no simple growth trend emerges,with volumes fluctuating between a low of 167 million tons in 1992and a high of 186 million tons in 1998. In real terms, this latter repre-sents approximately 99 000 million ton-kilometres of transport activi-ty-

What Figure 1 indicates is that traffic growth has been driven by twodominant and relatively limited activities - the transport of iron orefrom the Sishen area in the Northern Province to Saldanha on the Orexline, and more powerfully by the carriage of coal from mining areas inMpumalanga to Richards Bay. As noted above, these two niche activi-ties are of world class, both technically and in user-cost terms. What isseriously less than world class is virtually everything else in the railarena. The category of general freight (a congealed aggregate of activ-ities on routes other than the Coallink and Orex operations) fell fromover 130 million tons in 1976 to some 120 million tons by 1990, andthen more steadily thereafter to reach 95 million tons in 1999. Thisrepresents a 21 per cent decline between 1990 and 1999. Although thebroad category of general freight includes a number of low-value pri-mary products, its secular decline is part of a longer-term phenome-non, whereby rail has lost command over higher-value, and conse-quently higher-rated, goods.19 This also has some serious implicationsfor pricing and sustainability.

In terms of legislation,20 public rail has historically been recognised asthe national domestic carrier, and as such has been required to act as a

19. G.D. Van der Veer, The role of rail transport in the total transport scene: Freighttransport. Annual Transportation Conference Proceedings, Council for Scientificand Industrial Research, Pretoria, 1982, p 6

20. Republic of South Africa Constitution Act, Act N° 32 of 1961, section 103(1)


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

229

common carrier and to operate several uneconomic freight (and, par-ticularly in the past, passenger) services. Moreover, the same legisla-tion required the national transport services to operate in terms of busi-ness principles, and to balance their books. The latter was always in-terpreted very broadly, and was applied to the public transport sector,not to individual modes, and certainly not to discrete services withinmodes. The result was a complex web of cross-subsidisation, withgenerally profitable ports and pipelines bailing out loss-making rail ac-tivities.

Within rail, the tariff philosophy adopted was that followed by mostcommon carriers (in all transport modes) who face a diverse trafficbase covering a broad spectrum of commodity types and values. Thatstandard practice is to raise rates on higher-value cargo, in respect ofwhich demand is relatively price inelastic, in the process charging"what the market will bear", while low-value cargo, for which demandis conventionally believed to be more price elastic, attracts lowerrates.21 This type of "Ramsay" pricing has been a permanent feature ofSouth African railway rating, but in the last quarter of the 20th Centu-ry, and particularly in the 1990s, it has become increasingly difficult topursue, for four principal reasons. The first was the increase in low-rated bulk traffic associated with the new bulk-export ports. The sec-ond was the increased dominance of containerised general cargo traf-fic, for which unitary tariff rates became the only sensible pricingmechanism. The third emanated from the recommendations of a seriesof public Commissions of Inquiry and multi-disciplinary studies thatwere tasked to investigate domestic transport policy.22 Common to allof these was the advocacy of a more competitive domestic transport

21. In the case of high-value goods, freight charges represent a small fraction of totaldistribution costs, hence cargo owners are relatively unresponsive to tariffchanges, and are consequently vulnerable to price mark-ups. For low-valuestaples, the opposite argument applies.

22. The most influential included the Schumann (1964), Marais (1969) , and VanBreda (1973) Commissions, and the broadly-directed National Transport PolicyStudy (NTPS) group, orchestrated by the Department of Transport in the mid-1980s.


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

230

market, characterised by freer competition between road and rail, withboth modes charging the "right" (cost-related) prices for their services,but with rail relieved of the financial burden of providing uneconomicservices, most notably the transport of commuters from distant Blackdormitory suburbs to industrial areas. The fourth and most powerfulreason has been the emergence of a largely unregulated road freighttransport industry based on prices that are emphatically "wrong", inso-far as heavy freight vehicles systematically underpay for the use of theroad infrastructure. The result of these influences is a set of rail pricesthat more closely approximates underlying cost, albeit based on capitalcost levels that are below long-term capital replacement needs, butwhich nonetheless fails to undercut road rates,23 and which thereforealso fails to stem a migration of general cargoes from rail to road.

There is ample evidence that these institutional factors have been aid-ed and abetted by rank bad service to general users, and towards theend of the 1990s also to bulk-cargo clients, on the part of rail itself.Moving South Africa reports widespread dissatisfaction with generalfreight services24 and unreliability in the case of breakbulk freight.These views are also strongly reinforced by a study of the perceptionsof cargo owners and port-ancillary operators in the port of Durban,who indicated substantial unhappiness with operating performancemeasured by such indicators as rail wagon availability and supply oflocomotives at the times that these services were required. These tren-chant criticisms were not limited to the general cargo arena, butspanned across a broad gamut of neo-bulk and low-value bulk com-modities, to a point where the viability of some export shipments wasjeopardised.25

In sum, the combination of indifferent service quality and an inabilityto set prices at levels that are both competitive and sufficient to cover

23. Department of Transport, op cit, p 5124. Ibid, p 4625. Ethekwini Municipality, "Durban Metropolitan Local Economic Study - Report

5: The Port of Durban - Characteristics, user perceptions and growth constraints",unpublished, 2003


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

231

long-term capital replacement needs, bodes ill for the sustainability ofthe general freight services of rail in South Africa. Public utterancesby senior officials and recommendations of transport economists sug-gest that rail should be fulfilling a larger role in the long-distance car-riage of commodities other than bulk mineral exports. Without a fun-damental revision of modal costs and general quality of service deliv-ery, it is hard to see how these fine sentiments can be transformed intochanged behaviour on the part of transport users and rail transport pro-viders.

4. ROAD TRANSPORT AND INTERMODAL COMPETITION

The dominant symbol of change in South African transport marketsduring the decade of the 1990s has been the absolute and relativegrowth of road transport, to a point where it is now the largest singledomestic mode of freight transport. This transformation has been ef-fected from a narrow base, for prior to the 1980s, commercial roadhaulage operators were regulated virtually out of existence by a permitsystem that systematically favoured rail on long-haul inter-city routes.Some significant road haulage was effected by private companiestransporting their own products, often over short distances; but publiccarriers were responsible for a minuscule market share.26 By 2000,road had entrenched itself as the dominant land mode in both nominaltonnage terms and in terms of "real" transport activity. At the sametime, the regulatory and policy framework within which road transportfunctions has lurched from draconian dirigisme before the 1980s tovirtually untrammelled laissez faire by the 1990s. There is very littlepositive that can be said about either arrangement, and much that therecent (non-) policy environment has done to destroy equitableintermodal competition, and to threaten the survival of the economy'sroad infrastructure. The story of South African road freight transport isconsequently largely a story of regulation, its relaxation and its mis-

26. Van der Veer, op cit


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

232

management. The road hauliers themselves are almost incidental ac-tors in the drama.

Until well into the 1980s, the commercial road freight industry re-mained a stunted creature, denied natural growth by the legacy of the1929 Road Motor Competition Commission, whose recommendationsto control road transport in order to eliminate "wasteful and destructivecompetition" were embodied in the 1930 Motor Carrier TransportationAct.27 This piece of legislation provided the blueprint for the contro-versial permit system that effectively shut commercial public carriersout of the long-distance land transport market.

Some relaxation of these restrictive practices followed from the provi-sions of the 1977 Road Transport Act and its subsequent regulations.These extended the gamut of permit-exempt goods, increased the radi-us of unrestricted carriage, and lifted permit restrictions in certain ar-eas. This did not signal the end of the road permit system, and it didnot immediately fashion a competitive land transport market in SouthAfrica, but it did usher in a period of somewhat greater competitionbetween road and rail. Formal deregulation of road transport was onlyto come in 1989 with the abandonment of the permit system, but there-after liberalisation of the environment in which road freight transportoperates proceeded at breakneck pace. The principal changes from thelate-1980s onwards are:

• In 1987, a Government White Paper stipulated that a comprehen-sive Road Transport Quality System (RTQS) should be in place be-fore transport deregulation took place. Maximum permissible vehi-cle lengths were increased to 20 metres, increasing GCM to 47 tonsand maximum payload to 33 tons.

• In 1989, the road permit system was abandoned, signalling de factoderegulation of road transport, but without the RTQS being imple-mented.

27. Peter Freeman, The recovery of cost from road users in South Africa - Part 1:Background and theory, National Institute for Transport and Road Research,Pretoria, 1982, p 27


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

233

• In 1990, maximum permissible vehicle lengths were increased to22 metres. Interlink multi-trailer combinations come into regularuse for long-haul freight traffic.

• In 1991, GCM was increased to 56 tons, permitting a maximum (le-gal) pay load of 37 tons.

• In 1994, overloading tolerance of five per cent was allowed onGCM, increasing effective pay loads to 39 tons for lightweight trail-er combinations.

• In 1996, axle load limits of 8,2 tons, for which most South Africanbridges and roads are designed, were replaced by a limit of ninetons, with a five per cent tolerance. Legal payloads of up to 45 tonscould be achieved.28

From the end of the permit system in 1989, in the space of a singledecade, the South African road freight transport industry has movedfrom the periphery of the market to a position where it operates withsome of the largest heavy vehicles in the world that are not restrictedto designated routes.29 Under these circumstances, it is not surprisingthat road's freight market share has risen markedly, largely at the ex-pense of rail. Precise estimation of modal shares is bedevilled by apaucity of accurate data relating to road freight activity, most notablyactivities of private carriers transporting their own products. Unlikethe port and rail sectors, no continuous time series of performance ex-ists, so more fragmentary sources must be relied upon. One suchsource is the work of van der Veer, whose data show private "inhouse" carriers controlling some 35 per cent of total land transport ton-nages in 1980/81, while the share of public carriers stood at some 12

28. Railroad Association of South Africa (website www.rra.co.za), 2001; WillemKempen, "The road to ruin", Financial Mail, 19 January 2001

29. Maximum permissible Gross Combination Mass per vehicle in South Africa of 56tons (or 58,8 tons with the five per cent overload tolerance) exceeds that in otherSADC countries except Zambia, and stands well above levels in developedeconomies. In the United States the federal limit is 36,7 tons, although higherloads are permitted in some States, in most European countries it is 40 tons, andJapan and the United Kingdom permit a GCM of 38 tons.


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

234

per cent in a still heavily-regulated environment.30 This contrastsstrongly with 1999 estimates produced by Spoornet and the RailRoadAssociation, which put total land-based transport volumes at some 820million tons, of which rail controls 180 million tons (or 22 per cent ofthe total), and commercial hauliers 140 million tons (17 per cent),leaving 500 million tons (or 61 per cent) in the hands of private carri-ers of their own products. All of these exercises suffer from the com-mon weakness of a nominal tonnage rather a real ton-kilometre basis.This has the effect of over-stating road's share in general and the shareof generally short-haul private road in particular.31 By the late-1990s,rail transport generated some 99 000 million ton-kilometres of "true"transport activity, while a very rough and ready estimate puts road'soutput at approximately 113 000 million ton-kilometres.32 This sug-gests that road is now the senior partner in the South African landtransport market, with a modal share of about 55 per cent of aggregatetransport activity, while rail commands the remaining 45 per cent, al-beit with a traffic base heavily skewed towards basic bulk commodi-ties.

This represents a startling transformation in the fortunes of road trans-port, and one that would be both welcome and unremarkable if it cameabout as a consequence of freer-working and more efficient transportmarkets, driven by the right price signals. Sadly, the overwhelmingweight of evidence points to a market mechanism so distorted by ex-ternalities that the resultant pattern of resource allocation threatens thesustainability of both the road infrastructure and the rail system.

South Africa's road infrastructure is substantial - comprising a littleunder 60 000 kilometres of paved roads - but by no means all of this isin good condition. Moving South Africa reports only 18 per cent of na-

30. Van der Veer, op cit31. Average rail haul is approximately 550 kilometres, perhaps comparable to long-

haul road. Private road movements are much shorter, generally of the order of 100kilometres or less.

32. RailRoad Association of South Africa, op cit


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

235

tional roads to be in "very good condition".33 Inter-city trunk roads aregenerally in the best condition, and long-distance secondary and ruralroads in the worst. The biggest problem is the long-term financialsustainability of the road network, with a backlog of R40-57 billion"just to maintain the current-sized network - already deteriorated - ina steady state".34 At the heart of this problem is the absence of clearuser charges for the country's roads. Users pay through the fuel levy,through licence fees and through charges on toll roads; but with theexception of the latter there is no direct link to specific road expendi-ture. In many areas, but particularly in the case of heavy road freight,the "prices" that are in place are the wrong prices, thereby "distortingthe true economies of road freight with effects on modal choice andthe balance of modal competition".35 Many studies have explored theextent to which various users pay for the damage they inflict on theroad infrastructure, and although the precise results generated are opento interpretation, most reveal a strongly positive correlation betweenvehicle mass and the extent of under-payment. A study commissionedby the Automobile Association in the mid-1990s indicated that lightvehicles overpay by some seven cents per vehicle-kilometre, or byR3,02 billion annually (in 1994 prices) in aggregate terms. All catego-ries of heavy vehicles underpay, with the extent of under-recovery ofpavement damage rising from 15 cents per vehicle-kilometre in thecase of 2-axle heavy vehicles, to Rl,49 for 6-axle heavies, and to apeak of R2,04 per vehicle-kilometre in respect of the heaviest 7-axlerigs. Total under-recovery from the heavy road freight industry wasestimated at R3,87 billion in 1994.36 These estimates are based on le-gal maximum payloads, and would rise more than proportionately inthe case of overloading. The obvious conclusion to draw from theseexercises is that licence fees and road levies faced by light road usersare too high, but this overpayment is more than offset by an array of

33. Department of Transport, op cit, p 7934. Ibid, p8035. Ibid, p 4936. Automobile Association of South Africa, Heavy vehicle overloading in South

Africa, Durban, 1995


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

236

"prices" for heavy road users that are manifestly too low. These wrongprices are passed on to cargo owners in the form of freight rates thatdo not capture full economic costs, and consequently distort patternsof demand.

What this analysis suggests is that developments that took place prin-cipally during the 1990s have resulted in a situation where too muchfreight transport activity in South Africa is conducted by road, in vehi-cles that are too heavy, and which in the process inflict significant netdamage costs on the roadway. This does not imply that heavy roadhauliers are villains (other than those that systematically overload), butthat operators and users alike respond to the wrong price signals. Theimmediate task of road-freight policy-makers is to get prices right.

5. CONCLUSION

The overall picture of the South African freight transport sector thathas taken shape in this account, is of a system that is large and sophis-ticated by regional standards, but one that has coped increasinglypoorly with several fundamental requirements during the 1990s.

The South African ports are the maritime giants of the Southern Hemi-sphere, and produce certain products - notably bulk services - to thehighest world standards. Unfortunately, these are not the port servicesthat stand highest on the economy's agenda. More important is en-hanced capacity and lowered user cost in respect of general cargoes topower manufacturing output and exports, yet it is precisely these portfunctions that have largely been overlooked since 1990 as candidatesfor significant new infrastructural investment. A realignment of portpriorities will consequently be needed as a matter of some urgency.There is, however, reason for confidence. Basically profitable ports aregood candidates for privatisation, and future private port entrepreneursare likely to be responsive to the rents offered by growing general traf-fic.


Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

237

The outlook for rail and road-rail competition is more complex andless sanguine. Rail has, to an increasing extent over the last decade,exhibited the same asymmetry as the ports, with strength in bulk oper-ations and weakness in general freight. There is spare capacity on arte-rial routes, which poor operating performance partially negatives. Railcapillaries serving lower-density areas face a cloudier future, as theseare poor candidates for concessioning to private operators, unless thelatter can be relieved of financial responsibility to maintain the rail in-frastructure. More ambitious models of infrastructure/operating sepa-ration, along the lines already initiated by the ports, may be a possibili-ty, but a rail renaissance cannot realistically be fashioned without a si-multaneous reconsideration of the entire regulatory and pricing envi-ronment within which land transport functions. Moving South Africacalls for a "restoration of value-based competition between rail androad", with a larger stake for rail in the long-haul transport of a widearray of products.37 Such an outcome would be excellent news for theSouth African economy, but it would require a firm commitment onthe part of policy-makers to re-establish the right price signals in trans-port markets. Without such commitment, a better alignment betweentransport performance and national objectives may prove elusive.



Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

7:19

12

Nov

embe

r 20

14

the freight transport sector in the 1990s

Documents