The Port Of Durban From An Economics Perspective Economics Essay

3.1 Introduction

This chapter will examine the Port of Durban from an economics perspective and will seek to expand on the general theory presented in the literature review and apply it specifically to the Port of Durban. This chapter will also serve as a foundation for the proceeding chapter which will analyse the various CBA options and data for Durban. The ports significance and impact will be examined in the context of the South African and local economy through its income and employment generating effect. Though the quantity of cargo moving through a port is important, of more interest is the type of cargo that a port focuses on.

3.2 The South African Port Sector

Before examining the Port of Durban in isolation, it would be prudent to briefly discuss the South African Port scenario in a broader sense. In South Africa, ports are considered national assets and are managed by the government run recently by SAPO. South Africa is a major sea-trading nation comprising of approximately 8 trading ports, namely, Durban, Richards Bay, East London, Port Elizabeth, Mossel Bay, Cape Town, Saldanha and the under construction Coega. South Africa has evolved into a major sea-trading nation over the last four or so decades and in 2002 handled 3.6% of world sea trade by volume. In terms of ton miles or real activity, this figure increases to 6% of global trade, placing the country within the top 12 globally and resulting in a global maritime activity share that is more than 20 fold its global GDP share. Sea trade constitutes more than 90 percent of trade in South Africa and ports play a critical social and economic role both nationally and regionally. The majority of the port activity is concentrated on the east coast of South Africa. A stark illustration of this fact is that Durban and Richards Bay together make up 76% of sea trade in the country. Traffic growth in the 1990’s was derived from two primary regional points and sources, namely Durban from a general cargo perspective and Richard’s bay from a raw materials perspective. Richards Bay, which deals primarily in bulk goods, such as coal, ore and steel, has seen its annual tonnage increase from 55 million tons in 1989 to in excess of 90 million in 2000. Viewing perceived value in terms of tonnage is a flawed approach since in terms of economic linkages and value-adding, handling a ton of coal is not the same as handling a ton of refined goods. The figure below illustrates the breakdown of sea trade activity by port in South Africa. It can be seen clearly that Durban and Richards Bay are giants in comparison to the other ports. (Chasomeris, 2003 and Jones, 2002)

Fig 17: Total Traffic Volume in South Africa

Source: Department of Transport, 1998 and Jones, 2001

The South African Ports sector experienced significant capital intensive investment in the 1970’s and 1980’s, which was biased towards the bulk shipping sector. However, world trends have seen a migration towards containerisation and unitisation and South Africa is no exception, with the country utilising containers for the first time in1977. Up until 1990, the available capacity could cater for national traffic levels of approximately 1 million TEU’s level. The lack of adequate container capacity, combined with growing demand, brought with it a multitude of problems. On the demand side, South Africa became a democracy and re-entered the globalised world, resulting in a noticeable rise in seaborne container volumes, due to liner carriers returning to the South African trades and increased trade liberalisation. The upsurge in volumes produces inevitable negative consequences of delays and vessel queues. By 2000 the combined amount of annual TEUs handled in South African ports was 1.8 million and this was encompassed using with the same basic container quays that had been constructed in 1977. There was some limited capital investment in strategic area’s in the 1990’s, such as cargo extensions to bulk and neo-bulk facilities in Richards Bay. The new millennia brought with its bolder and more ambitious port investment initiatives. A new industrial hub status port in the Eastern Cape, which was earlier envisioned but never actioned upon, was now being constructed. Secondly, the Durban general cargo infrastructure has received significant upgrades and extensions such as extensions to landside facilities as well, deepening and extending cargo handling superstructure and infrastructure as well as deepening and widening the harbour entrance. Because of the age and mismatch of the cargo handling infrastructure, productivity has lagged that of international levels, resulting in congestion that is a constant feature of local ports. There were also supply side issues to deal with such as liner route becoming more specific and centred around hub status ports. As such, hub status ports have to provide capacity that exceeds national demand, making attainment of hub port status difficult in capacity constricted scenarios. This is compounded by the reluctance of ship-owners to migrate shorter routes such as Port Elizabeth in South Africa. South African ports relative competitive stance with their southern hemisphere counterparts can be gauged from the table below. Looking at both indicators, South African ports emerge as clear leaders on both the African and Southern Hemisphere front. Richards Bay is ranked first on the table in terms of total traffic, as it has a large amount of coal and other bulk cargoes passing through its doors. Durban, although ranked 3rd overall, is ranked 1st in the container category it is clear that Durban is the leading multi-purpose port in South Africa and the Southern Hemisphere. (Jones, 2003; Jones, 1997; Department of Transport, 1998 and Lawrence, 2000)

Figure 18: African and Southern Hemisphere Port Traffic

Port

Total Port Traffic

(m tons)

Rank

Container

Traffic

(TEUs 000s)

Rank

Richards Bay

91.5

1

5

15

Newcastle

73.9

2

9

14

Durban

49.7

3

1291

2

Santos

43.1

4

945

4

Sydney

24.6

5

999

3

Melbourne

22.3

6

1322

1

Casablanca

19.8

7

311

9

Abidjan

14.6

8

434

7

Auckland

13.3

9

561

6

Cape Town

11.8

10

395

8

Lagos

9.1

11

1782

11

Mombasa

8.9

12

219

10

Buenos Aires

7.8

13

716

5

Dakar

7.2

14

149

13

Port Louis

4.7

15

161

12

Source: ISL, Bremen, 2001, Jones 2003 (Selected ports, 2000)

3.3 History of the Port of Durban

The port is situated on the east coast of South Africa at coordinates 31o 02’E in longitudinal and at 290 52’S in latitudinal terms. Trading activities in the port of Durban can be traced back since 1824, with the port quickly gaining a favoured status among seafarers amd traders due to it being a natural harbour. Interest in Durban Bay grew tremedously in the early years of its operations, with imports doubling between 1849 and 1850. This, coupled with larger vessels, resulted in a much needed expansion to the harbour entrance. Over a century later, Durban has 63 berths and 6 repair berths, which can be broadly seperated into five main segments of the port. The first segments has two piers and has a multipurpose function thats handles general, parcel and unitised cargo. The second segment of the port is located by Salisbury Island and Island View. A third segment is the Maydon Wharf area, which contains private terminals as well as terminals controlled by Transnet. The Point terminal area and the Bayhead area’ are the fourth segment and fifth segment respectively. Below is a picture of the port of Durban that illustrates the five segments discussed.

Figure 19: The Current Layout of Durban Port

Source: Google Earth, 2010

3.4 Economic Significance of the Port of Durban

As, can be seen in figure 17 above, the logistical strength of the national shipping infrastructure, rests primarily in KZN. The port of Durban, like all other public ports in South Africa, is an example of a port under national jurisdiction, its official name being the National Ports Authority (NPA), thereby allowing centralised planning. Durban is a port of choice because of its infrastructure in place enabling it to be a full service general cargo and container port . In addition to this, durban is well serviced by an adequete rail and road infrastructure, which links it to the economic hub of South Africa, Gauteng. In addition to this, the KZN region is a large economic region in itself and is second only to Gauteng in South Africa. Figure 21 below, illustrates a snapshot of the South African port sector for 2009. In terms of total cargo tonnes handled, Durban has 20% of the market and is dwarfed by Richards which has more than double Durban’s tonnage handled, at more than 40%. Richards Bay, which was constructed in the 1970’s, has had an enormous impact on Durban’s port planning and functions. The primary reason for its existence was to serve as high-mass export point for raw materials such as coal. Richards Bay also diversified its goods base to include, at a lower cost, goods types that were traditionally the domain of Durban such as neo-bulk cargo like steel, alloys and forest type products. At the time of Richards Bay construction, Cape-sized bulk vessels were too large to enter Durban. (Jones, 2003 and Stats SA, 2010)

Figure 21: Port Cargo and Vessel Statistics in South African Ports

 

RICHARDS BAY

DURBAN

CAPE TOWN

SALDANHA BAY

TOTAL SA PORTS

Durban as a % of Total

TOTAL CARGO HANDLED:

77,631,154

37,419,282

3,058,601

56,475,625

182,735,369

20%

 

 

 

 

 

 

 

GENERAL CARGO VESSELS:

247

705

220

373

1,648

43%

BULK VESSELS:

1257

930

320

921

3,603

26%

CONTAINER VESSELS:

42

1883

897

784

4,233

44%

TANKERS:

184

646

159

344

1,542

42%

VESSEL TOTAL:

1874

4848

2440

3489

15,879

31%

 

 

 

 

 

 

 

TOTAL TEUS HANDLED:

6,273

2,395,175

1,382,052

NA

4,334,612

55%

Source: NPA, 2009 (Note table has been edited)

Looking again at figure 21 above, it can be observed that even though Durban lags other ports in gross tonnage of cargo, it still has by far the most number of vessels docking. One of the major reasons for this was the emerging dominance of Richards Bay, which forced Durban to concentrate on lower-volume bulk, break-bulk and liquid-bulk. This enabled great diversity within the port in terms of cargo type as well vessel type and quantity. Additionally, vessels that carry break bulk are traditionally far smaller than that of traditional bulk, thus explaining why more vessel docking are in Durban than Richards Bay for the same amount of cargo ceterus paribus. With reference to the figures above, it can be observed that Durban has 43% of total general cargo vessels, 42% of total tankers and 44% of total container vessels. The most important figure, in relation to Durban, is that of TEU’s handled since this is where its dominance and significance come to the fore. Durban has the ideal structure to handle containers and since Richards Bay has inadequate structure for containers, Durban’s dominance in containers was from the outset. Jones (2003) show that a growing international trend of shipping lines with regards to containers is to organise trade and activities around so called “hub” ports which meet and cross at “sub-regional transhipment nodes”. This arrangement is biased for the existence of a single hub type port on the eastern shores of the Southern region of Africa. Since, Durban is the country’s major container port, is well frequented by major shipping lines, has terminal and hub status, it is quite reasonable for it to remain South Africa’s primary container port. The other alternatives on the eastern sea board are not really competitors when it comes to containers. Richards Bay is primarily a bulk port and does not have the adequate infrastructure to extend its activities beyond this scope. Maputo has large deviation costs from traditional shipping lines as well as limited depth and capacity. Port Elizabeth has weak land side links to Gauteng as well as having limited local demand to justify a major port there. (Suykens, 1984; Jones, 2001 and Jones, 2003)

Even though Durban lags Richards Bay in terms of pure tonnage, this in itself is a poor yardstick of economic impact and significance since no account is taken of cargo value or employment propensities of infrastructure required. Generally, in terms of economic and employment impacts, general cargo provides the most followed by dry-bulk cargo and lastly liquid-bulk. Bearing this in mind, comparing two ports only on the basis of tonnage is frivolous and more specifically in Durban’s case it can be seen that from a ports perspective, it handles higher valued cargo than Richards Bay. This is especially evident when one considers one job is created per 47000 tonnes of cargo handled at Richards Bay, whereas in Durban, one job is created per 7500 tonnes of cargo handled. Figure 22 below further illustrates the economic richness and opportunity that containers present. Additionally, in 2004 an average container vessel spent R2.94 million per port call, far exceeding the R1.8 million for a breakbulk cargo vessel as well as exceeding the R1.3 million for a bunker vessel. (Suykens, 1984; Jones, 2001, Tempi, 2006 and Jones, 2003)

Figure 22: Port of Fremantle’s Economic impact by Cargo Type

Cargo Type

Output ($m)

Value Added ($m)

Household Income ($m)

Employment (no.)

Direct Effects

 

 

 

 

Containers

177

121

73

1331

Other General Cargo

45

30

18

340

Liquid Bulk

35

20

8

158

Dry Bulk

83

44

25

459

Other

1

1

7

Total

341

215

124

2294

 

 

 

 

 

Direct + Indirect Effects

 

 

 

 

Containers

382

240

125

3195

Other General Cargo

96

59

31

800

Liquid Bulk

67

38

17

441

Dry Bulk

181

100

50

1339

Other

2

1

1

19

Total

728

440

223

5792

Source: Bureau of Economic Transport Economics Australia, 2000

As is the case with South African ports, the port of Freemantle in Australia, shown in figure 22 above, derives the most economic prosperity from containers from both a direct and indirect perspective. Even though containers account for only 13% of activity in the port, they contribute 55% to economic activity. Consequently, containers have the greatest employment generating effects, followed by dry bulk and the liquid bulk. Though dynamics differ from port to port in terms of infrastructure, administration, socioeconomics and geography, a broad consensus can be reached from the figure above encompassing a kind of “rule of thumb” approach. As such, containers offer the most economic opportunity for a port and since Durban already focuses on this area, it would be prudent to continue with this trend. Thus, it is quite evident that both the present and future comparative advantage of Durban port rests in the realm of containerised cargoes due to reason shown above. Also, since the port is so aptly designed for and dependant on containerised cargo, the removal of this great economic magnifying source would be particularly devastating on the Durban region as a whole. (Jones, 2001 and Jones, 2003)

Looking at figure 23 below, it can be seen that the Durban port has seen an extraordinary increase in containers, with annualised growth of between 8% and 10% for the last decade. As was shown above, containers form an integral cog in the Durban port machine from an economics and social perspective since they provide a source of trade, income and employment. Container growth has been driven by a range of factors such as rising volumes of world trade and reduced trading barriers, the migration of cargo to containers from other handling systems, South Africa’s improved economic performance and rising per capita incomes. The facets examined below are containers landed, shipped and empty and as the diagram shows, all three categories have increased from 2002-2007. The growth between 2002 and 2007 is nothing short of spectacular, but this growth has not come without costs and constraints. However, needing containers and providing adequate space for them are two entirely different things and this will be explored below. Also, we have seen that general cargo is the richest form of cargo and has the largest employment benefits. South Africa needs extended general cargo capabilities and in this respect, Durban’s needs are similar to national needs. It is thus clear that Durban needs the container industry for continued survival and prosperity, but whether the container industry needs Durban as much remains to be seen. (Jones, 2003)

Figure 23: Total TEUs Landed, Shipped & Transhipped

Source: NPA, marketing graphs, 2008

Durban’s greatest strengths, namely its ideal location, good economic linkage and strong infrastructure, have also evolved to be its Achilles heel, since its popularity especially for containerized cargo, has seen demand surge amidst mostly fixed infrastructure. With the growth of sea trade demand, the real problems of Durban are the lack of adequate marine infrastructure, but its role as port with terminal capacity, and the managerial capacity and willingness to operate the present container terminal at acceptable performance levels. A supply side response by the authorities to these demand pressures has been slow and limited. The growth of containerised cargo volumes has put the port’s container terminal under sustained pressure since the mid-1990s, and at times has overwhelmed available capacity. The consequences of which have been frequent queues of container vessels, unduly high berth occupancy rates, and delays to container vessels and their cargoes. The port area is inundated with industrial and commercial development, making space an expensive premium, above all for neo-bulk space intensive cargoes like steel and forest products. It is therefore no surprise to see certain of these cargoes migrating to Richards Bay, where space is at less of a premium. The Durban-Gauteng rail line possesses substantial spare capacity, but operating problems associated with the availability of Transnet have reduced the reliability of rail. This problem is particularly serious for certain bulk terminals that are reliant on rail since for bulk commodities rail is the cheapest and most efficient form of transport. Previously, Durban’s major economic disadvantage was its inability to host Panamax size-threshold ships due to its lack of depth. However, after recent capital investments, the entrance width has been increased from 110m at its narrowest to 220m and the depth in the outer channel from 12m to approximately 19m. However, this is far from adequate and as can be seen in Ircha (2006) which states that hub status type ports must have the following in order to remain relevant:

• Container-stacking densities of 2000-4000 TEUs per hectare;

• Sustained ship-to-shore gantry crane productivity of 50 moves per hour;

• Three day dwell times;

• 30-minute truck turnaround times;

• On-dock rail service; and

• Water depths by the berth of 15 metres and more.

Currently, Durban subscribes to one of these parameters, and if it wishes to become efficient and remain productive and relevant, authorities should try to subscribe to all of them. Doing so would require significant capital investments such as infrastructure expansions. (Lawrence, 2000; ISL, 2001; Fairplay, 2003, Ircha, 2006, Transnet, 2010 and Jones, 2003)

3.5 Multiplier Model

The theory of the Keynesian multiplier was covered quite extensively in the literature review. Figure 22 above touched on the multiplier process for the port of Freemantle, but the concept will now be explored and applied in far more detail. The economic impact of port activities on the local economy can be subdivided into three broad areas. The first area is that of directly port-related or port generated activities, that would cease to exist if the port were to close. The second area is that of indirectly port-related activities and pertains to backwardly-linked services and infrastructure. The third and final broad category is termed induced effects, and is in fact the multiplier effect from other inputs. It arises as those employed in the previous two categories, re-spend their money in the local economy, thereby increasing the original economic impact. Jones (1998) conducted a study so as to ascertain the Port of Durban’s economic impact on the local economy. Figure 24 below is taken from that same study and as can be observed, 24 000 direct port related jobs from approximately 360 businesses are created through first round inputs. Of the 24 000 jobs, approximately 8500 are from Transnet, which is an indication of the significant role that the institution plays in the local region. The 24 000 figure translate into a wage bill of approximately R950 million rand in 1994 wage level. Assuming an inflation rate of 10% per annum, this figure would equate to approximately R4 Billion in 2010 terms! Coupled to this, many port activities were in fact excluded from the above calculation such as insurance, financial services, medical services and legal services. (Jones, 2003)

Another reason why the employment figure is conservative is that it fails to account for the induced or multiplier effect. As shown in the literature review, the economic or employment effect is extended far beyond the initial spending impetus whereby the final round of total expenditure normally far exceed the initial input. The multiplier varies from region to region depending on the average marginal propensity to consume, taxes, and how much money is kept within the local region. Jones assumes that since the majority of port employees are in fact low to middle income earners, which is not an outrageous assumption. Bearing this in mind, an average tax rate of 20%, MPC of 0.85 and a retention rate of 0.85 is used to formulate the multiplier value. The data is substituted into the multiplier equation from the literature review and yields a multiplier value of 2.4. The port of Seattle conducted an economic impact analysis and depending on which assumptions they used, the multiplier ranged from 2.9 to 4.4. The port of Lake Charles Harbour also conducted an economic impact study and used a multiplier of 2.6 and the port of Hastings derived a multiplier of 1.58. Thus, the figure use by Jones is in no way over the top when one looks at other port economic impact papers and it even falls on the lower end of the spectrum. The box below illustrates the calculations that were used to obtain the multiplier. At 1994 prices total income generated by the port is approximately R2.3 billion. Once again, if we assume a 10% increase per annum, in 2010 price terms, this would equate to R9.6 Billion! (Jones, 2003; Meyrick Associates, 2007 and Martin Associates, 2007)

Figure 24: Multiplier for Durban (1994 prices)

α = 1

1 -c [(1-t) r]

Substituting the various values

= 1

1 -0.85[(1-0.2)0.85]

=2.4

Calculating Equilibrium income for wages only:

Yo = αA

Yo= 950 X 2.4

= R2.3 Billion

Calculating Equilibrium income for all expenditures:

Yo= (950+500) X 2.4

= R3.5 Billion

Source: Jones, 2003

Even with the multiplier effect, the regional economic impact of the port is under estimated since wages and salaries are not the only costs in a port. Industries which provide inputs and services to port establishments are excluded. In the same paper, Jones attempts to calculate these very costs and some of the examples include paper, ropes, cranes, hooks and property costs. Jones does this by working out that on average 48% of total costs are non wage costs and based on this assumption, a 1994 figure of R500 million is generated from port related expenditure which is not linked to wages. This amount extrapolated to regional labour elasticity’s, induces a labour figure of approximately 7000 jobs. The refineries around the port employ around 1800 people and the Island View area about 500 as well. Thus, as Jones rightly says, the port and port related activities generate around 40000 jobs in the local economy, a figure which eThekwini online concurs with. Looking at the box above, it can be calculated that the total economic impact of the port is R3.5 Billion in 1994 prices. In 2010 monetary terms, this equates to roughly R14.62 Billion. Additionally, eThekwini online states that the port and related industries contributes over 20% of Durban’s GDP and approximately 1.5% of national GDP! Thus, it is quite evident that the port and its related clusters are integral to the Durban community in terms of employment and social stability. (Jones, 2003 and www.thekwenionline.org.za, 2010)

Figure: 25 Durban Port Employment and Output (all data at 1994 levels)

Industry/Sector

Number

Employment

Wage bill

 

 

 

(R mill)

 

 

 

 

Portnet

1

5400

240

Portnet dredging

1

112

6

Spoornet

1

3217

115

Terminal operators

11

2213

90

Liquid bulk terminals

3

275

16

C&F agents

138

3600

135

Ships agents

37

1350

65

Ship chandlers

17

400

ns

Container depots

3

366

13

Container parks

7

260

ns

Container logistics

3

140

6

Shipowners & operators

5

11002

ns

Ship repairers & builders

5

9603

34

Stevedores

24

1650

45

Cargo equipment suppliers

2

200

ns

Road haulers

>75

15001

ns

Bunker services

2

110

5

Offshore services

3

80

3

Tallying services

5

1204

ns

Security

3

3001

ns

Marine contractors

2

114

5

Customs & Excise

1

300

ns

Other State

3

1001

ns

TOTAL

>360

23867

~R950

Source: Jones, 2003

3.6 Constraints to Expansion

As shown in Figure 18 above, Durban is the largest general cargo port in Africa and the second largest in the southern hemisphere, and Durban being a port city will benefit from any growth in international trade volumes especially of the general cargo type. Although Durban’s port infrastructure is extensive, at present it suffers from critical capacity limitations. The port currently provides 63 berths that can be used for cargo related activities as well as repair facilities for a further 8-9 vessels. These capacity constraints are encountered in respect of the port’s marine infrastructure, cargo-working facilities and its overall articulation with landside cargo distribution systems. The constraints are indicated in the figure below, which illustrates the situation for Durban in 2004/5, considering that the teu amount was 2,395,175 teus for 2009, it becomes clear how grave the capacity situation is. Considering how grave the capacity situation is, it is indeed surprising that only short term capital investments have been undertaken over the last two decades. Towards the end of the previous century, there were some capital extensions such as gantries, larger container areas and straddle carriers. In 2002, more gantries were added as well as 20 straddle carriers. The second part of the 2002 project was the relocation and specialisation of areas within the port, namely pier 1. All these short term improvements will result in the port having a present day capacity of 2.5 million TEUs. Already in 2005 the container terminal were operating at 90% capacity and now 5 years hence, with TEU’s handled being 2.4 million in 2009 or 96% capacity, there is a pressing need for Durban to increase and improve its container handling operations. (NPA, 2009 and Muller, 2004)

Figure 26: Port of Durban Capacity Constraint

Terminals

Current traffic M ton

Theoretical capacity M ton

Spare Capacity

Percentage used

Bulk Liquids

23,800,000

 Unlimited

Unlimited

 –

Motor vehicles units

171,365

220,000

48,635

77.89

Coal

1,800,000

2,500,000

700,000

72

City

2,400,000

5,200,000

2,800,000

46.15

Containers

1,724,218

1,900,000

175,782

90.75

Break bulk

4,200,000

6,300,000.00

2,100,000

66.67

Total excl vehicles

33,924,218.00

16,120,000.00

5,824,417.00

 

Source: NPA, 2006

Though this paper views the port from an economics perspective, it must be borne in mind that this is only one of the uses for the port. The table below, taken from Tempi 2006, illustrates some of the other uses of the port of Durban. These include disease management, recreation, food production, bird refuge and conservation. Thus, there is always an opportunity cost for any form of expansion in the port and this must be balanced against developmental conditions of the region. Ports which are located in urban areas as is the case of Durban, have a great public demand to access the port for leisure based activities such as sailing, fishing, paddling, water skiing and canoeing. However, if the land around the port has become a trendy place to frequent, the other developments such as restaurants, nightclubs and shopping areas would also bid for the valuable land space. Besides port development directly competing with these activities for space, the already established developments can impose a negative externality in terms of noise, traffic congestion, negative aesthetics, property devaluation and pollution. Applying the above to Durban, one can see that the significant point developments fall quite clearly in this category. There are many conflicting activities happening simultaneously in the port of Durban and the recreation versus commercial activities is one of them. A prime example of this is the development of a Multi billion rand waterfront adjacent to the harbour that includes luxurious property developments, an aquarium, a water world and various water sports. Thus, by implementing infrastructure to develop the port to handle further cargo, an opportunity cost is created in the form of lost recreational activities.

(IEM, 2006 and Ircha, 2006)

Figure 27: Port of Durban Services

Source: Tempi, 2006

Referring again to the figure above, it can be seen that the environment and ecological role of the port is quite significant. Before the Durban Bay was converted to a harbour, it was a pristine estuarine lagoon that was abundant in wild life, plants and swamps. From the 1800’s onwards, after the port became a popular throughput destination, the port gradually ceased to be a natural haven though dredging and construction. Thus, even though there are still plentiful bio organisms in the port, this amount pales in comparison to what the bay held in its prime and as such there is increasing environmental pressure to not do any further damage to the natural habitat. Most of the original sand banks have been lost through dredging, resulting the remaining sandbanks been considered quite essential because of their essential nutrient contribution to the neighbouring biomass as well as their rarity and existence value. A factor contributing to their existence value is their rarity in terms of this type of habitat been only available in Richards Bay and Durban Bay. A key feature of the sandbanks is their stability which is created by the contribution which microscopic organisms and macro benthic invertebrates make through enhancing resilience and increasing surface area of the sandbank. A higher surface area from the sandbank benefits the bay via the filtrating mechanism that it provides, which ensures that the water entering the bay is somewhat cleansed. Besides the sandbanks, there are also the heritage site status mangroves of which approximately 15 hectares are remaining in the bay. These are vital to the bay for biodiversity and sustainability and also share the same body of water as the grasslands and sandbanks, thereby participating in the energy transfer properties of the water medium. There are numerous activities that depend on the proper functioning of this ecological area, with the most obvious been the angling industry, be it for subsistence or recreational purposes, which resulted in 16000 angling outings in 1987. The second activity is that of bait collecting which had a harvested amount of 2.5 tonnes in 1995. Bird watching and academic research are also frequent activities which depend on a well functioning ecological system. Forbes and Demetriades (2007) show that the water area of the bay has been reduced by 57%, the sandbank by 86%, sea grass by 100%, bird species by 94% and the mangroves by 97%. The complexity of the ecosystem means that a simple correlation between habitat and species loss is nonexistent. Figure 27 below illustrates the ecological losses of habitat and bird life for selected time periods. Figure 22 illustrated on an overhead satellite photograph the location and extent of the sandbanks and mangroves. (Forbes & Demetriades, 2006; Mander, 2007 and IEM study, 1996)

Figure 28: Port of Durban Habitat Loss

Source: Tempi, 2006

An ecosystem by definition is interlinked and as such changes in one part of an ecosystem can have drastic consequences in another part. Deeper and wider channels, brought on through expansion, could have a detrimental effect on the sandbank and via interlinked systems on the entire ecosystem. This would be further exacerbated by larger post panama whose more powerful propellers would cause disturbances in the ecosystem as well. According to the environmental section of the Tempi process, there is no feasible scenario under which the partial or total removal of the sandbanks can be justified. Additionally, the option of rehabilitation, relocation or replacement is stated to be a poor substitute for the natural sandbanks as the sandbank type and quality is stated to be quite unique. This is firstly due to the fine sediments of the silt canal and secondly, the physico chemical nature of the ecosystems in the bay. The silt canal in the bay is directly affected through the inflow of freshwater via rivers thereby influencing salinity, oxygen levels and turbidity. This in turn determines the amount and ration of the various micro and macro organisms in the bay and replicating this balance in another location would prove near impossible. The 1996 IEM study concurs with this view and shows that removing the sandbanks for infrastructure purposes would have a detrimental effect on the ecosystem as well as 60 hectare loss of water area. The IEM study also describes the possible negative externalities that would come from port expansions such as traffic congestion in Glenwood and the Berea, the compromising of capital investments on Victoria Embankment and the obstruction of sea views in surrounding areas. (IEM, 1996 and Tempi, 2006)

The above constraints are absolute constraints in terms of physical obstacles that may hinder expansions. An often overlooked constraint is that of efficiency lost or relative constraints. In a survey conducted in 2003 by Jones, various firms were asked for comments and opinions with regards to this very issue. The results were that not a single respondent viewed the terminal and cargo handling services provided by the port as good. Additionally, rail services were seen in the same light, which is a worrying occurrence considering how crucial rail is to bulk services. The survey also showed the lack of faith in management’s competence with 64% stating the lack of necessary skills in the port as a major problem. Thus, in a way, the inefficiency of the management structures is itself a constraint on the ports expansion and even current workings. This has come into the fore recently because the situation is further exacerbated by the capacity constraint of the port. (Jones 2003)

3.7 Growth Forecast for Durban Containers

It has been shown above that Durban is South Africa’s foremost container port. The containers that are handled at Durban port make up 65 percent of all containers in South Africa and the port of Durban, having terminal status, is by far the best adapted port for this function. It has also been shown that though containers offer the best return on investment in the form of economic linkages, but Durban is suffering from severe capacity constraints. As shown previously, the port management’s response to these capacity constraints has been slow, limited and short sighted. The port is experiencing congestion problems in the container area currently but the more significant problem is the expected growth of container volumes as has been stated. To accurately predict what kind of additional capacity is needed for the future, one must first calculate the actual cargo volumes for the related periods through some form of demand forecasting. Econometric forecasting is estimating a quantitive value about the probability of a future even occurring using underlying fundamentals’ and can be broadly divided into three categories, namely, cross sectional, time series and panel data. Cross sectional data analyses a group of data points at a single moment in time. For the purposes of forecasting, time series econometrics is the most conventional approach, and this is forecasting using data taken across time intervals. Panel data is a combination of both cross sectional and time series data and involves analysing a specific group of data over a period of time In order for an econometric model to be meaningful; it must be relevant, rational and significant. Achieving these requirements means model and data selection are of the utmost importance and this selection depends on available resources. Time series models vary according to data availability and manipulation and some of the more common models are autoregressive integrated moving average (ARIMA), autoregressive distributed lag (ARDL), vector autoregressive (VAR), auto regressive conditional heteroscedasticty (ARCH) and generalised autoregressive conditional heteroscedasticity (GARCH). Looking at time series in detail is beyond the scope of this dissertation, but the first three models listed above will be briefly examined as they are the most common ones used. After which, we will briefly look at some container forecasting models for the port of Durban. (Gujarati, 2003 and Stopford, 1997)

The ARIMA model is based on two components, namely an autoregressive component and a moving average component. The autoregressive component means that a current value, say Y, is dependent on a previous or lagged value of Y. The moving average component relates to how the error term and its lagged values contribute to the dependent variable, say Y. The term integrated related to how many times the model must be differenced in order to become stationary. In time series, a model is considered stationary if its mean and variance are constant over time and the covariance value is dependent on the distance between the two time periods and not on when the time periods are. If a model is not stationary, the regression results will be spurious and inaccurate. See below an example of an ARIMA model which has a constant, a lagged value of Y and a lagged value on the error term to help explain the outcome of Yt. (Gujurati, 2003 and Muller, 2004))

The ARDL model has two parts to it as well, an autoregressive component which was explained above and a distributed lag component. The distributed lag component is simply lagged independent variables that are used to explain the dependant variable as well as lagged values of the dependent variable. When a model uses both present and lagged values of its explanatory variables, it is a dynamic model. The reason that lags are used is that many economics reaction take time to be processed in an economy, an example of which is the interest rate impact in South Africa which takes approximately 18 months to be felt in the real economy. There are three main reasons for lags. The first is the psychological reason and relates to habitual behaviour, fear, optimism and perception. The next is the technological reason and involve the use of capital and technology in the economy. The last is institutional reasons and to contractual obligations. See below an example of an ARDL model where Y is the dependent and explanatory variable, U is a constant term, X is the independent variable and E is the error term. (Gujurati, 2003 and Muller, 2004)

The VAR model is atheoretic, meaning they do not have any theoretical underpinnings, and uses simultaneous endogenous variables only. In all time series models, to a varying extent, the past data and relationships between data flows is extrapolated to the future. It is this very condition that sometimes leads models to be entirely wrong and example of which is the current world recession which no model predicted. However, by including a big enough data set with lagged GDP, previous recessions should be included. Predicting any future any future event with certainty is impossible, but models help by assigning probabilities, and this is why econometrics is often considered more art than science. The results of some econometric times series studies for the port of Durban’s containers demand as well as the actual number of TEUS for selected years are illustrated in the figure below. A key structural problem, which none of these models address, was that between the years 2007-2009, the worlds encountered its greatest recession aka “the credit crunch” since the 1930’s depression. As such the actual number of TEUS in 2009 is actually less than the 2007 figure. Muller (2004) uses an ARDL, ARIMA and VAR model to calculate future container volumes for the port for the period 2004-2013, however since the ARDL had the lowest errors, the figures from that model were used. The 2006 and 2007 predictions are underestimated, whereas the 2009 figure, probably due to the non-prediction of the recession, are overestimated. Floor and Van Niekerk (2001) use a straight linear equation that depicts the number of TEUS as a function of GDP. Their prediction runs from 2000-2020 and is clear from the comparison with the actual TEUS that their model is not very accurate. This is a clear example of how more dynamic models such as those above is more suited for time series type data and predictions. The official model used by the Tempi process and Transnet has been undertaking by Graham Muller and Associates. The model predicts that that by 2013 Durban will have almost 5 million TEUS passing through it, whereas the Muller model predicts that this figure predicts the 3.75 million TEUS for the same year. (Gujurati, 2003; Floor and Van Niekerk, 2001; Muller, 2004 and Transnet, 2006)

Figure 29: Comparative results of Forecasting Models

Comparative Years

Muller (2004)

Transnet (2006)

F&V (2001)

Actual

2005

1843403

1898483

1553661

1898483

2006

2014839

2198523

1639112

2202841

2007

2202220

2485731

1729263

2480223

2009

2772436

3189633

1924713

2395175

2013

3751631

4928918

2384382

NA

2020

NA

8280853

3287688

NA

Source: Floor and Van Niekerk, 2001; Muller, 2004 and Transnet, 2006

The model used by is forecast until 2036 where the expected TEUS will be in the region of 17 million TEUS. The near future consequences that will follow will be congestion, increased costs and delays. In the long term, this could even lead to Durban losing its dominant container status.

Figure 30: Durban Container Long Term Forecast

Source: Transnet, 2006

3.8 Conclusion

Thus the cost and opportunities of increasing port capacity has many perspectives, conflicts and problems. Questions of development and sustainability as well as environmental parameters all contest to become the dominant perspective. From the above, it is clear stop gap short term measure are not going to solve alleviate pressure for very long and as such large expansion options must be considered if Durban is to remain relevant. Durban’s ever increasing port congestion and so a long term solution needs to be in place if Durban port is to remain internationally competitive. This is where Durban is at a crossroads in terms of where to locate an additional container terminal. The theory of CBA was covered in the literature review and it is through this methodology that this dissertation will attempt to answer these fundamental questions.