One ongoing challenge within rail relates to a strategic decision in the late 19th century to build a narrow-gauge railway system. The carrying capacity of railway wagons, which is the biggest driver of marginal costs, increases with a wider rail gauge. In most cases, carrying capacity could at least be doubled by using a wider standard rail gauge (KTH Railway Group 2017).

This problem is compounded by a gross vehicle mass (GVM) for road freight vehicles in South Africa that is significantly higher than that of developed countries. South Africa’s maximum is 56 tonnes versus Europe’s 40 tonnes (EPRS 2011) and the USA’s 36 tonnes (Shuster 2015:6). The upshot of these two factors is that a South African rail wagon can carry just two twenty-foot equivalent units (TEUs), whereas a truck can carry three TEUs.

This has contributed to a shift of freight from rail to road. However, with overloading, road conditions have worsened, and with the fact that many trucks are not roadworthy, this has led to increased accidents and vehicle inefficiencies, among other externalities (Furter 2014).

Rapid deregulation accelerated the loss of general freight from rail to road (Havenga et al. 2021:1). That contributed to a loss of railway density and reduced the contribution that was essential to cover the fixed costs of moving freight by rail. In turn, this put pressure on tariffs and lowered funds available for maintenance.

In a vicious cycle, this pushed even more freight from rail to road (Havenga et al. 2021:2). This was exacerbated by the government’s reluctance to invest in rail corridor services that could have preserved freight on the main rail corridors and aided in assuring rail competitiveness relative to intense competition from road in a deregulated market.

Total rail-friendly freight was estimated at 181 billion tonne-km in 2019, whereof about 141 billion tonne-km was transported by rail (FDM™ 2019). While bulk rail transport rates have been found to be competitive with that of international peers, general rail freight transport rates are not (FDM™ 2019). In 2019, most of the tonne-km were attributable to the large volume of export coal and iron ore on rail. Rail delivered only 18 billion tonne-km of general freight, compared to the estimated 47 billion tonne-km it should have delivered. That makes for a gap of some 30 billion tonne-km. Similarly, palletised long-distance freight is negligible. This is shown in Figure 1 and Table 1. As customers have increasingly resorted to road transport, congestion at the harbours has increased.

The ‘export mining’ freight-flow segment refers to bulk export mining that is transported via rail over long distances, with high density and limited origins with the export ports as destinations. The ‘domestic mining’ segment refers to bulk mineral mining for domestic processing, usually stockpiled at manufacturing plants. These flows are more complex and have medium to long transport distances, which involve more common-use facilities and sharing of road and rail infrastructure with other value chains. ‘Semi-beneficiated’ (i.e. semi-processed) freight flows are related to heavy break bulk that requires specialised wagons (such as mining commodities transported to value-added processing centres or semi-beneficiated goods transported between value-added processing centres). These high-density flows occur between plants, or plants and distribution centres (DCs), with multiple points of origin and a few destinations (typically DCs).

These flows span long transport distances nationally, which can be short in metropolitan areas. ‘Finished palletised goods’ are the flows of final processed goods between DCs, usually on pallets. These flows are high density and have complex supply chain management requirements but few origin-destination (OD) pairs over long transport distances nationally, with short metropolitan transport distances. The ‘agriculture’ segment refers to low density extraction flows of agricultural goods to cities or production centres with many OD pairs over short to medium transport distances. This segment also includes the delivery flows of agricultural goods in the opposite direction with the same characteristics. Furthermore, the ‘agriculture’ segment includes rural interchanges, which are flows between farming areas that are low density and seasonal. The development of freight flow segmentation is based on research conducted to determine all flows in South Africa and categorise these to enable macrologistics planning (Havenga 2012; Havenga & Simpson 2018b).

Expanding on what is presented in Table 1:

Competitive forces have also been constrained. The vertically integrated structure of the rail sector has meant that there were incentives for Transnet to prevent the entry of private operators. This allowed Transnet to keep competition out of the system, which would have served as a market-driven motivation to streamline operations and allocate investment to their most efficient uses.

Transnet has been underspending on maintenance for a long time. According to an analysis undertaken by the Africa Rail Industry Association (Zote 2022), Transnet Freight Rail (TFR) spent R3.4 billion on maintenance in 2012. If maintenance spending had kept pace with inflation, TFR would have had to spend R5.57 billion in 2021–2022. In fact, it spent only R2.6 bn. Moreover, a significant portion of the spending earmarked for maintenance is being diverted to pay for security, as vandalism and theft run rampant (Dludla 2022).

Underinvestment in rail infrastructure has exacerbated poor operational performance. Based on a review of Transnet’s annual reports, there has been a lack of investment by Transnet in rolling stock technologies and locomotives, resulting in an ageing and unreliable fleet requiring higher maintenance levels. No new locomotives were procured between 1994 and 2008, and few new wagons were procured. On the contrary, the government simultaneously invested significantly in road corridor infrastructure through the South African National Roads Agency SOC Limited (SANRAL), a government agency ‘responsible for the management, maintenance and development of South Africa’s national road network’ (SANRAL 2022). South Africa’s road freight sector also invested in road trucks and trailers over the past 15 years (Figure 2).

Transnet infrastructure has been damaged, vandalised and stolen (Larkin 2022). Cable theft and malicious damage to signals, wayside equipment, locomotives and even theft of rail tracks have taken place. Replacement is impeded by slow procurement processes. The flooding in KwaZulu-Natal during June 2022 damaged rail infrastructure (DPE 2022). The consequence is key rail lines carry speed restrictions, and in some places, only a single rail line is operational, increasing freight transit times.

Transnet is the primary owner, through the Transnet National Ports Authority (TNPA), of port infrastructure at the nine seaports for which it is responsible. Transnet Port Terminals (TPT) has a 100% market share of car terminals. Between 2010 and 2016, TPT increased its market share for container terminals from 97% to 98%. During the same period, its breakbulk market share declined from 78% to 69%. It is also responsible for about 50% of bulk terminals (Chasomeris & Gumede 2022:55).

Being both the owner of port infrastructure and an operator creates an inherent conflict of interest. Transnet is incentivised to maintain a monopoly in port operations rather than allowing the entry of private competitors. It has been largely successful at keeping this status quo in place despite private multipurpose terminal operators being well established. In this regard, Transnet continues to maintain a virtual monopoly over the container and car terminals. This militates against efficiency and innovation.

While state participation in the ports sector has continued in South Africa, many countries have ‘privatised their ports, introduced private competition into the ports or facilitated state-owned ports companies to become internationally competitive’ (TIPS 2014:4). Bichou (2021:23) has shown that the country’s container terminals’ regional and global peers saw a gradual improvement of 30% in yearly combined average terminal efficiency scores (based on intertemporal data envelopment analysis [DEA]) from 2010 to 2019, compared to a decline of 15% in South Africa. South African terminals were outperforming their peers until 2014 but have been losing ground since then. Operational efficiency is an additional drawback. Compared with an international norm of loading about 35 units per crane hour, in 2019, Transnet was only achieving 20–25 units per crane hour.

Transnet National Ports Authority’s ownership monopoly of South African ports has resulted in port charges that are among the highest in the world, despite the recorded lower efficiency. While benchmarking by the Ports Regulator of South Africa (PRSA) showed that charges for bulk commodities are competitive, charges for containers and automotives are twice as expensive as at other ports (TIPS 2014:55). Inefficiencies add further costs, for example, additional berthing costs while in harbour. Because Transnet operates a complementary rather than competitive port model, the relatively uniform prices charged at South African ports do not reflect the actual costs. There is cross-subsidisation within the ports, between the ports, and between the ports and the railway. Additionally, ‘ad valorem wharfage charges’ were used to address a large pension fund deficit that Transnet inherited from its predecessor when it was corporatised, which distorted port prices (Stratton 2003:256). This has undermined the sustainability of the port system.

The conflict of interest should have been addressed through the corporatisation of the National Ports Authority, as provided for in the National Port Act of 2005. However, this was not implemented, based on resistance from Transnet, who have argued that vertical separation would negatively affect transport efficiency. This resistance has been supported by the Department of Public Enterprises (TIPS 2014:6).

Concerns about lacking levels of competition, cross-subsidisation and efficiency caused by the vertically integrated, monopolistic structure of the ports informed the establishment of the PRSA in 2007. The PRSA’s primary role is to provide economic regulation and equitable access in the South African ports sector. Transnet National Ports Authority has, therefore, been required to apply to the PRSA annually for tariff increases since the financial year 2010–2011.

The historical pattern of underinvestment in the ports continues (Bichou 2021:8, TIPS 2014; Whitehouse 2022). Investment has been limited because of Transnet’s need to cross-subsidise rail. Actual investment by Transnet has repeatedly been lower than the budgeted capital expenditure. There have been delays in a number of critical port projects, which include the deepening and lengthening of the berths at the Durban container terminal. Delays on critical projects add to the operational challenges.

It is evident from Transnet’s recent audited financial statements that the company’s financial performance has been deteriorating over the past 5 years (Transnet 2021, 2022b). Whereas previously it had realised net profits, it recorded a loss in 2021. This was attributed to a substantial increase in costs that occurred at the same time as a decrease in revenues. Throughout the 5-year period, Transnet has had a current ratio, i.e., short-term (current) assets over short-term (current) liabilities, of below one. This means Transnet has insufficient capital to meet all its liabilities maturing in the short term. This deteriorated further in 2021. The leverage of the company’s balance sheet has been gradually increasing, and in 2021, the company’s poor profitability implied that the level of debt was very high compared to the company’s earnings and cash generation ability, a situation that remains. This is also reflected in the deteriorating interest cover and cash interest cover.

Transnet has six operating divisions, namely, TFR, Transnet Engineering, TNPA, TPT, Transnet Pipelines and Transnet Property. Transnet Freight Rail is by far the largest division of Transnet, generating a revenue of R37.8 bn in 2022 (47% of Transnet’s total revenue), followed by TPT (18%) and the TNPA (18%). However, TFR only accounts for 44% of the EBITDA (Earnings before Interest, Tax, Depreciation and Amortisation), whereas TNPA represents 30%. This indicates that effectively, cross-subsidisation is taking place from the ports to the rail to maintain the overall financial stability of Transnet, but this is resulting in underinvestment in port infrastructure (Transnet 2022b:20).

While port revenues have been increasing, rail freight revenues have been declining, which has had a significant negative impact on TFR’s profitability. It reflects the impact of the fall in volumes being transported by the division. TFR’s volumes have fallen from 215.1 million tonnes in 2018-2019 to 173.1 million tonnes in 2021–2022 (Transnet 2022b:26). The declining performance of the rail division has meant that funding available for maintenance and capital expenditure shrank from year to year.

Up until 2021, Transnet’s financial statements were qualified, as the company was unable to report irregular expenditure fully and accurately, much of which related to corrupt transactions, including the 1064 locomotive procurement. The total irregular expenditure from 2011 to 2021 amounted to R104 bn (Transnet 2021:16).

In 2022, rating agencies placed Transnet on a negative credit watch, in advance of its USD 1 bn bond redemption that became due in July 2022 (Moody’s 2022; S&P Global 2022). The rating agencies considered Transnet’s liquidity position weak. Delays in finalising the financial statements and obtaining an unqualified audit opinion have made it difficult for Transnet to access capital markets for refinancing, and even when they had done so, appetite for Transnet bond issuances has been tepid, with investors mainly interested in short-term debt. The average duration of Transnet’s debt has fallen significantly.

For the first time in over 20 years, the government announced that funding was to be allocated to Transnet. During the 2022 Medium Term Budget Policy Statement (MTBPS), the Minister of Finance announced that R5.8 bn would be allocated to Transnet to repair infrastructure damaged by flooding and to maintain the locomotives (National Treasury 2022:28). However, far more funding and continued support from government will be required if Transnet is to invest meaningfully in its infrastructure.

Through the voluntary severance programme implemented by Transnet during and after 2020, critical skills have been lost. Virtually, the entire executive committee is made up of people new to Transnet. Critical operational employees have also been lost, including the heads of five of the six divisions, the chief operations officer of the coal line and the chief of planning (Paton 2022a).

Transnet was one of the SOEs most deeply and negatively affected by state capture, including the procurement of 1064 locomotives at highly inflated prices. Corruption had become deeply entrenched within the organisation, so much so that, for a period in 2019, procurement came to a virtual halt because of the number of supply chain employees who had to be suspended, while investigations were underway to confirm fraud, corruption and irregular expenditure (Paton 2022a).

As described in 2021 research by Havenga et al. (and quoted where applicable in this section), freight flows move from the place of extraction or production to the place of utilisation or consumption, which result in key flow patterns. In turn, these flow patterns bring about five overall freight flow segments. In each of these cases, the role of rail can be defined, and the impact of disruptions or failure to deliver the rail services can be identified

Exports of coal, iron ore and manganese:

The movement of local minerals to domestic beneficiation centres:

The flow of FMCG commodities of higher value between manufacturing facilities, distribution centres and retailers:

Flow of semi-beneficiated commodities between intermediate and final processing facilities:

The flow of agricultural bulk from rural areas and delivery of consumer goods to these area:

It should be noted that last-mile freight (which should always be transported via road) and certain ring-fenced movements (such as pipelines, conveyor belts and buoys for fuel movements in ports) are excluded from this segmentation.

For this research, the direct cost impact of rail disruptions and failures on value chains is calculated based on the following:

For both ports and rail, the following costs were not quantified:

Apart from the direct cost, this research also provides an extensive economic analysis using multipliers to better illustrate the wider implications (i.e. the induced cost) of the macrologistics failings for the South African economy. While various other sources of multiplier literature were consulted (Batini, Eyraud & Weber 2014:10; Janse Van Rensburg, De Jager & Makrelov 2022; Ramey 2019), the multipliers employed in this article were largely informed by gross domestic product (GDP) multipliers for South Africa (split by sector) found in 2013 research by Burrows and Botha. Although their research only presents multipliers between 1980 and 2010 in 5-year intervals, more recently, Schröder and Storm (2020:11) stated that the average trends seen in Burrows and Botha’s work still hold true. Desktop research conducted for this article also found no evidence of other GDP multipliers for South Africa to refute this. More specifically, this research used the published type II GDP multipliers, which quantify the impact change the components of GDP (namely, government spending, consumption and net exports) has on total GDP within specific sectors. Therefore, within a sector, a change in any of these components relates to a change in total GDP via the multiplier (Burrows & Botha 2013).

A conservative assumption was made to quantify the impact of the reduced export volumes. It was assumed that government spending and consumption would stay constant within the coal mining sector since the 2013 research by Burrows and Botha, with the impact being quantified by the reduction in net exports (which is directly measurable through the paper’s methodology). In South Africa, all mining activities generate between 0.5% and 7% royalties to the government, depending on the price and financial position of the mineral producer. The reduction in export volumes would directly translate to a reduction in royalties for the government which means less ability for the government to spend without borrowing money. The assumption to keep government spending constant is, therefore, conservative for two reasons. The first reason is that if the government spending of a portion of GDP within the coal mining sector remains constant, it has a neutral effect on the economic impact. The second reason, which is intertwined with the former, is that the impact of a government budget shortfall is also not accounted for. As the South African government borrows more money through selling government bonds, the government’s debt increases, and it needs more money to service the debt, which leads to more borrowing. This devalues the government bonds, which forces the government to increase the interest rate as it is seen as a higher risk investment, and on its current path, it is creeping towards a default if it cannot increase its revenue generation capacity or reduce fiscal spending. Refer to ‘Figure 7.1’ in the 2023 Budget Review for more information (National Treasury 2023a). Furthermore, the domestic government bond interest rate has gone up from 5% to 7.6% from 01 January 2022 to 31 January 2023. The increased cost of debt to the South African government also puts downward pressure on government spending across all sectors, which is excluded from this analysis.

Given this context, a multiplier of three is assumed for both the absolute loss of volumes and the opportunity cost of higher transport costs. The former’s multiplier is based on the rounded average historic multipliers of the ‘coal mining’ and ‘other mining’ sectors, while the latter’s multiplier is based on the rounded average historic multiplier for the ‘transport and storage’ sector (Burrows & Botha 2013:26). While these multipliers might seem significant initially, it is important to realise that if foreign exchange was earned by a mine, the mine could have paid more salaries, which would have entered the economy and had a positive knock-on effect. Alternatively, the mine could have bought other services that also enter the economy and have a positive knock-on effect. On the contrary, if the transport services were cheaper, buyers could have invested the differential elsewhere, which also would have had a positive knock-on effect. Along with consideration of these cascading impacts, it is also vital to remember the government debt dynamic explained earlier. The induced cost, therefore, includes both the direct and opportunity costs related to the inefficiencies of the ports and railway. Apart from using established work on South African multipliers to inform this calculation, the calculations in this article have been sense-checked by local macroeconomists. As these insights were provided on an individual and informal basis, the economists prefer to remain anonymous to avoid any sentiments being formally attributed to their respective organisations.

Considering all these factors, along with their knock-on effects on the rest of the economy, it is calculated that the direct loss to the economy that can be attributed to an underperforming railway increased from ~R28.2 bn or ~0.5% of GDP (approximately R5605.5 billion according to the World Bank) in 2019 to ~R152.8 billion or ~2.2% of GDP (approximately R 6793.4 bn according to the World Bank) in 2022 (Figure 3).

However, the direct cost, which measures the cost to service providers and/or freight owners for services, does not reflect the entire picture. Understanding the true extent of an underperforming railway’s impact on the economy requires a broader measure of the direct and opportunity costs, namely, the induced cost, to be calculated. The induced cost is estimated to be ~R84.5 billion or 1.5% of GDP in 2019 and ~R458.4 bn or 6.7% of GDP in 2022 (Figure 4).

Port disruptions and failures also have significant value chain cost impacts. Import trade value chains are negatively affected when production factors for the domestic industry are delayed. For example, delays in importing machinery used in agricultural or manufacturing processes have a significant knock-on effect. Export value chains also have negative delay time effects. Value chains are severely impacted when exports are obstructed, i.e., the opportunity for a sale is lost either because of capacity for market demand not being met or because of perishable produce losing value. Importantly, unlike for rail, alternative options are very limited when ports fail.

For the purposes of this research, the cost impact of port disruptions and failures on value chains is calculated based on the following variables:

Given the above, the calculated direct loss to the economy caused by underperforming ports (defined previously) increased from ~R9.7 bn (0.17% of GDP) in 2019 to ~R15.5 billion (0.23% of GDP) in 2022 (Figure 5).

Figure 6 shows the induced cost amounting to ~R29.2 bn (~0.52% of GDP) in 2019, which nearly doubled to ~R46.5 bn (0.68% of GDP) in 2022.

The estimated direct cost to the economy caused by rail and port disruptions and failures amounted to ~R37.9 bn (0.68% of GDP) in 2019 and reached ~R168.3 bn (2.48% of GDP) in 2022 (Figure 7).

The induced economic cost amounted to R504.8 bn in 2022 – which is nearly five times more than the R113.7 bn (approximately 2.03% of GDP) calculated for 2019. This means that between Transnet’s rail and port operations, almost R505 bn worth of disruptions and failures occurred in 2022, equivalent to 7.43% of GDP. See Figure 8 for a full breakdown of the various contributors.

The social impact of these disruptions, failures and losses is significant. Without these issues, 50 000 jobs could have been added in the mining sector alone (Paton 2022b). According to PwC (Ryan 2022), at an expected GDP growth rate of 1.5% per annum, employment – comprising both formal and informal jobs – will grow at a rate of 1.2%. This, in turn, means that a loss in GDP has a direct relationship with a loss in employment. In South Africa, about 14.5 million people are employed (Stats SA 2021:8). Using the relationship (Ryan 2022), a loss of 7.43% of the GDP could, therefore, be assumed to be the loss of 860 000 job opportunities (a conservative estimate), affecting the livelihood of many more South Africans. This also contributes to rising cases of protests and unrest and other social issues.

For benchmarking purposes, three global rail systems, with common characteristics and distinct raisons d’être, were examined. As described in 2013 research by Havenga et al., ‘North America (USA and Canada) has mostly privatised, large, vertically integrated railway companies that are highly productive and profitable’. The Russian, Indian and Chinese (RIC) railways are massive, vertically integrated, state-owned monopolistic railways that play an important role in socioeconomic development, industrialisation and infrastructure building (Havenga et al. 2013; Spector 2013). European railways have made significant progress with open access and have a strong focus on alleviating congestion and the negative environmental impact of road transportation. In 2020, North American and the RIC countries’ railways delivered 84.1% of all tonne-kms in the world. Adding all the European railways shifts this number to 85.9% (UIC 2020).

Tonne-kilometre indicators reflect the growing role of rail transport in world economies, in contrast to the reducing role of rail in South Africa (Figure 9). A country-level rail activity comparison confirms this reality (Figure 10).

Over the past 30 years, route kilometres in North America were rationalised by 40%, which led to skyrocketing tonne-kilometres, moderate growth in locomotive power and a 60% fall in employment (Figure 11, Figure 12, Figure 13 and Figure 14). The reduction in network size and relatively low tariffs allowed for two big developments: (1) locking-in customers and value chains and (2) allowing for value chain and supply chain designs that included the rail mode. Tonne-kilometres doubled in North America between 1980 and 2010, fluctuating around the 2010 level over the decade up to 2020. The figures give international comparisons with South Africa, and in the latter case, the export lines in these figures refer to the bulk coal and iron ore exports through Richards Bay and Saldanha, respectively.

In sharp contrast, RIC countries achieved a similar tonne-kilometre growth via an increase in route length, increased growth in locomotive power and a mere 15% reduction in employees. In Europe, tonne-kilometres have been static. While there has only been a moderate rationalisation of route kilometres, there has been a sharp drop in employment and locomotive power.

The international experience can be juxtaposed with South Africa, where its network size, i.e., route kilometres, has remained effectively unchanged. More specifically, less than 10% of South Africa’s route kilometres have been uplifted (FDM™ 2022), resulting in a heavy burden of fixed costs. There has been a significant rationalisation of employment, including skilled and experienced staff, leaving the network understaffed for its size, but still overstaffed for its output, adding to the high fixed costs. Locomotive power has also been reduced. Growth in tonne-kilometres is attributable to the coal and iron ore export lines.

These trends were driven by different reform agendas. North American railways were allowed to rationalise routes and staff, leading to private companies with low-cost bases, increased efficiencies and high densities. This allowed the retention of rail-friendly traffic in a deregulated environment, which helped attract private-sector investment.

In the RIC countries, the focus on rail as a tool for development is evident. Adding infrastructure was seen as a means to support growth. Through direct government involvement, rail has remained strong, despite high employment levels and lower tonne-kilometres than those in North America. Higher rail costs are tolerated because of significant transport distances and road transport being incredibly uncompetitive. Rail transportation is now growing faster than in North America, with the next big challenge being to insert these railways efficiently into supply chains (Lee & Shen 2020; NITI Aayog, RMI & RMI India 2021; Wagener, Aritua & Zhu 2020). For this, the drive for more private-sector involvement in the Indian and Chinese railways is currently underway (Cui, Pittman & Zhao 2021; Lohia 2023).

International benchmarking suggests that South Africa’s rail network is too large. Across the globe, there are around 1 million km of rail routes, 2% of which is in South Africa. Considering that South Africa contributes only approximately 0.4% of the world’s GDP (The World Bank 2022), this rail network is too large by comparison – it is estimated that it should be around 1%. While the stated intent behind the policy is to make Transnet profitable and lower the cost of doing business in South Africa, it has also been expected to contribute to development, and there has been reluctance from the government to allow Transnet to rationalise its rail network. Transnet first proposed this network reduction shortly after the election of the new government. The reduction would have led to job losses; however, a growing economy as a result of the reduction would boost employment significantly in the long term. This would not only offset the initial losses but also create more sustainable employment opportunities (Martin 2004).

The freight railway in South Africa is big enough (in terms of the relationship between the size of the economy and the size of the network) to emulate an American Class 1 railway. For this, the rail infrastructure and operations could remain vertically integrated, which could allow these to be restructured into a rationalised and freight-densified company to achieve improved benchmark indicators. Alternatively, it could be used as a social instrument, similar to what has been done in the RIC countries. As transport costs would increase initially because of increased fuel and environmental costs, such a strategic direction would require direct government involvement and substantial investment initially.

Because the closure of branch lines had a considerable impact on employment during the early 2000s (see Figure 14), rail subsidisation might be required to support further rationalisation of the network. Although trucks pay three to seven times more in toll fees than passenger vehicles do (based on various toll fees as published by Foresight Publications), this difference fails to reflect the disparity between their road impact accurately because estimates show passenger vehicles cause less than 1% of damage to roads (DoT 1997). Toll fees collected from passenger vehicles are, therefore, cross-subsidising the infrastructure damage caused by heavy freight vehicles considerably, with estimates suggesting this amounts to approximately R67 bn annually (which is essentially 41c per tonne-km). Because the average trucking fees for 2020 were around 94c per tonne-km, this translates to a passenger cross-subsidy of around 30% of total road freight cost. With the coal and iron ore exports excluded, the tonne-km market share target for rail is about a third of total freight; therefore, if rail fulfils its rail economic task, it should receive a third of the R67 bn in subsidies (approximately R22.3 bn) (FDM™ 2023). However, if rail transport is concentrated on the most economically routes, this subsidisation should not be necessary.

The World Economic Forum’s (WEF) Global Competitiveness Index (GCI) rankings can be used to assess South Africa’s maritime transport performance and competitiveness. South Africa’s ports’ quality was ranked 37th in 2017 and 2018, which is considerably better than its ranking of 48th in 2008. This ranking is also significantly better than the average ranking of 72nd received for overall transport infrastructure in 2018, especially considering the drop from being ranked 43rd in 2008. The country’s average road (down significantly from 38th in 2008 to 58th in 2018) and railway (down from 45th in 2008 to 47th in 2018) infrastructure rankings have also been poor compared to that of its ports (Figure 15) (WEF 2008–2018).

In Australia, large ports in Sydney are receiving attention to improve the integration between rail and ports to enhance trade. The ports are ‘investing in extra rail capacity to address long-term demand in a sustainable way’, boosting on-dock rail capacity to improve turnaround times, providing operators with more rail windows and allowing ‘more containers to be moved by rail to and from the terminal without the need for double handling’. A mode shift incentive scheme ‘to incentivise regional exporters to use metropolitan intermodal terminals for the last-mile journey and collection of empty containers is also being implemented’ (Thompson 2022).

Over the past two to three decades, the size of container vessels has increased dramatically, and these vessels can now carry loads of 25 000 TEUs (up from 5000 TEUs 25 years ago) (Jungen et al. 2021). South Africa’s container terminals were not designed to accommodate, load and offload vessels of this size. New models for the private sector and other investment models will have to be pursued to facilitate investment in new or upgraded container terminals to remain globally competitive.