A note on infrastructure quality in South Africa

Abstract

In South Africa the emphasis, both in research and in policy-making, seems to be more on the quantity of infrastructure than on improving its quality. This research note critically analyses the lack of quality indicators in quantitative infrastructure research, with the aim of redirecting attention to improving the quality of infrastructure in South Africa.

1. INTRODUCTION

Infrastructure is currently a highly debated topic in both international and local academic literature. In South Africa, various reasons are postulated for increasing investment in infrastructure. First, research into infrastructure investment, both local and international, suggests a positive (sometimes causal) relationship between higher infrastructure investment and higher economic growth (Calderón & Servén, 2004; Perkins et al., 2005; Fedderke & Bogetic, 2006). Second, South Africa is lagging behind the rest of the world in terms of a number of infrastructure variables (Bogetic & Fedderke, 2006). And third, this country will play host to the 2010 Soccer World Cup, an event that is an important catalyst for increasing government investment in infrastructure (Fourie, 2006).

Infrastructure investment seems to be the politically prescribed panacea for the South African economy. The reason for such investment seems to have shifted from primarily redistribution to encouraging higher economic growth. Under the auspices of AsgiSA, South Africa's accelerated and shared growth initiative, the 2007 budget has allocated R415.8 billion for spending on infrastructure over a 3-year period, which includes expenditure estimates by national and provincial departments, municipalities, public–private partnerships and public enterprises (National Treasury, 2007).

Yet the emphasis, both in research and in policy-making, still seems to be on more infrastructure rather than on an improved infrastructure. This research note critically analyses the lack of quality indicators in infrastructure research, with the aim of redirecting attention to improving the quality of infrastructure in its various forms in South Africa.

2. QUALITY OF INFRASTRUCTURE

2.1 Why infrastructure quality?

The distinction between infrastructure quantity, also called the volume of infrastructure (Calderón & Chong, 2004), and infrastructure quality is vague. Usually, infrastructure quantity is thought of as increasing the size of the infrastructure stock: building more roads and railways, extending the electricity grid or power-generation capability, adding more telephone lines or providing more household access to water and sanitation.

Infrastructure quality, however, is a catch-all phrase that includes all performance-enhancing improvements, both of the physical infrastructure and of the services provided (Fourie, 2006). The quality of transport infrastructure, for example, includes the condition of roads and airport runways, the timeliness of port services, and the safety of the railways. The quality of electrical utilities may be determined by the reliability of the electricity supply and the quality of the telecommunications infrastructure by the speed of connectivity.

Infrastructure quality therefore constitutes both the condition of the physical infrastructure and the performance of its services (Nordås & Piermartini, 2004). Maintenance improves the condition of physical infrastructure, while the performance of its services depends on the managerial capacity of its staff or the technical ability of its designers. As a clear example of the distinction, the value of a railway line between two towns depends on the condition of the actual railway and trains – maintained by means of regular safety checks and maintenance projects – and depends on the efficient running schedule operated by the station managers and train drivers.

Yet the distinction between infrastructure quantity and quality seems to be of no great importance to researchers: both are said to have a positive impact on economic growth. Admittedly some studies, in discussing the impact of infrastructure, argue for infrastructure of a better quality, although they implicitly refer to improving both the size and the performance of infrastructure stock (Calderón & Servén, 2004).

Not all infrastructure investment benefits society in the same way, however. The distinction between quantity and quality is important for understanding how investing in infrastructure leads to economic growth. Providing basic services such as water, electricity and sanitation (i.e. an increase in the quantity of infrastructure) would help alleviate poverty and promote redistribution (Calderón & Servén, 2004). To increase economic growth, however, investments in transport, telecommunication and electricity infrastructure are needed – not only to provide these services but also to improve them, which entails improving the quality of the infrastructure concerned.

Theoretically, investing in infrastructure achieves growth in two ways (Fedderke et al., 2006; Fourie, 2006): directly, since the cost of inputs into the production process is reduced; and indirectly, since productivity improves. While increasing the quantity of infrastructure may add to economic growth, maintaining the condition and performance of existing infrastructure may be critical to keeping up a high growth rate or, at least, to ensuring there is no sudden slowdown. An increase in the existing stock of infrastructure may lower the cost of inputs (building a new railway might, for example, lower freight transport costs, thereby reducing the input costs for firms) or may improve the productivity of the stock (an electricity connection might, for example, improve the productive capability of a rural sawmill). Deteriorating infrastructure (or inefficient service provision), on the other hand, may cause agents to cease production (e.g. as happened recently after the blackouts in major production centres across the country due to severe backlogs in the maintenance of power lines) or, at least, revert to the original infrastructure, which could lead to exorbitant costs, as the original infrastructure may not be able to handle the increased demand. The result of such reversion is most likely to be a slowdown in economic activity.

To investigate the impact of infrastructure quantity and quality on economic growth, Calderón and Servén (2004), using various measures of infrastructure and estimation techniques for a sample of 121 countries, find that both the quantity and the quality of infrastructure have a positive effect.1 For South Africa, using an array of infrastructure measures, Fedderke et al. (2006) find that infrastructure has a positive and significant effect on growth, through its improvement of labour productivity and raising of the marginal productivity of capital. A more recent study (Fedderke & Bogetic, 2006) confirms these results.

2.2 International trade

Various positive externalities are created through infrastructure investment, such as improvements in competitiveness, international trade and regional integration initiatives (Fourie, 2006). An increasing pool of evidence suggests that improved infrastructure quality has an important impact on international trade (Limão & Venables, 2001; Clark et al., 2004; Nordås & Piermartini, 2004), in essence because of the reduction in trade costs. According to Nordås and Piermartini (2004:5), trade costs can be divided into search costs, the cost of enforcing contracts, transport costs, tariffs and the cost of delays and uncertainties in delivery.

Infrastructure investment affects all such trade costs. Telecommunications infrastructure can reduce search costs and increase trade. Fink et al. (2002) find that including the cost of a telephone call in a gravity model has a significant negative effect on bilateral trade flows.2 Nordås and Piermartini (2004:5) argue that ‘the cost of not being able to place a telephone call or access the internet may be just as important as the cost of making the call’. They include a variable for telecommunications in a more sophisticated gravity model and find that it has a significant positive effect on the prediction of trade flows.

According to Nordås and Piermartini (2004:6), the quality of transport infrastructure is an important determinant of transport costs and the cost of time spent in transit. Hummels (2001) argues that, due to the substantial increase in air freight relative to sea freight over the past few decades and the increasing need for ‘just-in-time’ logistics, the cost of time spent in transit may exceed the transport costs – a view that is shared by Nordås and Piermartini (2004:6). Hummels (2001:21) finds, for example, that a 1-day increase in ocean transit between two countries reduces trade by 1 per cent for all goods and by 1.5 per cent for manufactured goods. Such an increase suggests that the average ocean voyage of 20 days implies a tariff of 16 per cent. Time is therefore an important competitive factor and hence also a trade barrier in its own right, according to Nordås et al., and it ‘not only affects the volume of trade, it more importantly also affects the ability of enterprises to enter export markets at all’ (2006:43). Port efficiency and transport networks connecting ports and industry are therefore crucial determinants of exports, not only of monetary outlays but also of the opportunity foregone to reach markets where fast delivery is essential.3

Uncertainty is another important trade cost. According to Nordås and Piermartini (2004), this uncertainty includes not only the cost of delays, but also the cost of the cargo not arriving in the condition in which it left (which implies insurance costs). A good-quality transport infrastructure – both of the physical infrastructure (condition) and the services it provides (performance) – can reduce the costs of uncertainty associated with trade.

Trade costs are a significant part of total trade. Hummels (2001) suggests that although the tariff effect of trade costs has shrunk over the past few decades – he estimates a reduction from 20 per cent in 1950 to 5.5 per cent in 1998 – transport costs have become steadily more significant in explaining trade patterns around the world over the past few decades. Therefore, since timeliness and reliability have both become very important factors in determining international trade, Nordås and Piermartini suggest that the ‘quality of transport infrastructure might have become a more important determinant of trade than in the past’ (2004:7).

A further effect of the quality of infrastructure on trade is described by Yeaple and Golub (2007), who find that the quality of infrastructure can help to explain the absolute and comparative advantages of countries through the impact of such quality on total factor productivities. Therefore specialisation in international trade not only depends on factor endowments but, increasingly, on the quality of public infrastructure provision (Yeaple & Golub, 2007).

Furthermore, while these benefits mostly accrue to exporting firms, lower trade costs also have substantial advantages for importing firms, both as lower input costs in the manufacturing process and as final goods for consumers. In fact, because of the high negative trade balance in South Africa, the initial benefits from higher quality infrastructure may be due to a reduction in the costs of imported goods rather than the dynamic gains of export development, which will only materialise in the medium to long run.

Increasing the size of the infrastructure stock will, of course, also reduce trade costs. Yet the quality of the infrastructure and the services that the infrastructure provides seem to be of equal importance in determining a country's ability to trade with the rest of the world, a factor that should be considered in any analysis of the impact of infrastructure investment.

Of course, international trade is not the only beneficiary of better quality infrastructure. Fourie (2006) notes several other important economic and social indicators that depend on the quality of infrastructure, including poverty, inequality and gender. In a panel study assessment of the impact of infrastructure on income distribution, Calderón and Chong (2004) find that both the quantity and the quality of infrastructure are important in reducing income inequality, with causality running from the former to the latter. Furthermore, their results suggest that quality issues are particularly important in industrial and emerging market countries and are relatively less important in poorer countries.

3. INFRASTRUCTURE QUALITY IN SOUTH AFRICA

3.1 Measuring infrastructure quality

Unfortunately, the effects of infrastructure quality are extremely difficult to measure. Perkins (2003) compiled data on South African infrastructure since 1875. However, none of the 19 infrastructure variables can be considered to measure purely the quality of infrastructure.4 Time-series studies using Perkins's data (including Perkins et al., 2005; Fedderke & Bogetic, 2006; Fedderke et al., 2006) are handicapped by the lack of adequate measures of infrastructure quality and may not represent the full benefit of infrastructure investment.

There is, however, some international evidence that gives an idea of how South Africa's infrastructure quality compares with that of the rest of the world. The data, compiled by Estache and Goicoechea (2004), show a number of infrastructural quality measures across countries (see Table 1).5

Table 1: Quality measures of South Africa's infrastructure performance

	Electric transmission and distribution losses (% of total output)	Telephone faults (reported/100 mainlines)	Time of commute in major cities (minutes/one-way trip to work)	Paved roads (% of total number of roads)
South Africa	8	48	35	21
Sub-Saharan Africa	19	57	34	25
Middle East and North Africa	14	23	25	56
South Asia	22	97	27	38
East Asia and Pacific	12	39	36	32
Latin America and Caribbean	18	24	29	36
Europe and Central Asia	18	34	29	76
Low income	22	64	33	30
Middle income	15	25	29	52
Upper middle income	14	18	29	57
High income	6	11	32	82
World	14	37	31	50
Source: Estache & Goicoechea (2004).

A more recent report by the South African Institute of Civil Engineering (SAICE) supports Estache and Goicoechea's (2004) comparative data. This report card, shown in Table 2, grades South Africa's infrastructure by awarding a symbol to each type to indicate the quality of its performance. A short summary of the performance of each sector is also provided. The qualitative evidence suggests that ‘although South Africa's built environment infrastructure is very good, even world class in parts, the relatively poor overall grade reflects extensive maintenance and refurbishment backlogs’ (SAICE, 2006:7). In most sectors, the condition of the infrastructure seems to be rapidly deteriorating, mainly due to a lack of regular maintenance, skills shortages and overuse. While overuse is in itself a consequence of rapid economic growth, combined with a relatively low increase in the quantity of infrastructure, maintenance and skills shortages (both technical and operational skills) are a result of poor planning, financing and education and training policies.

Table 2: South African infrastructure needs by type

Type of infrastructure	Grade	SAICE summary	Required investments
Water	D+ for Department of Water Affairs & Forestry infrastructure	Well-maintained but ageing bulk infrastructure reaching end of useful life, hence requiring refurbishment or replacement.	Increase in quantity.
	C+ for major urban areas	South Africa is one of the few nations where, in most urban areas, water can be drunk directly from the tap. Major ongoing improvements in provision of water and sanitation since 1994. However, erratic compliance with water quality requirements in most municipalities. Excessive water wastage (leakage). Shortage of skilled personnel.	Increase in quantity, especially in rural areas. Improvement in condition (maintenance to reduce leakage). Supply of skills to rectify deficiencies.
	D− for all other areas
Sanitation (including wastewater)	C− for major urban areas	Serious problems with management of many wastewater (sewage) treatment works. Excessive wastewater leakage and spillage, and frequent problems with on-site sanitation. Inadequate operation and maintenance capacity, and shortage of skilled personnel.	Improvement in condition (maintenance to reduce leakage) and especially performance (skills capacity).
	E for all other areas
Solid waste management	C− for major urban areas	Landfill sites in major urban centres well managed, but many municipalities, especially rural ones, have uncontrolled dumpsites, with attendant health risks. More widespread waste avoidance and recycling initiatives required.	Due to absence of infrastructural backlog, only investment in education campaigns.
	D for all other areas
Roads	C for national roads	Most in fair to very good condition, with recent strategic acquisitions in poorer shape. Increasing use of user-pays (tolling), but funding remains a challenge, especially given that key roads will soon require extensive refurbishing.	Improvement in condition.
	D− for all other roads	Generally inadequate funding and management systems leading to neglect of maintenance, combined with overloading, involving growing maintenance backlogs. Less condition monitoring than in the past. Shortages of skilled personnel. Decisions have been taken to stop maintaining some roads.	Improvement in condition (maintenance). Inadequate supply of much needed skills currently delaying maintenance.
Airports	B for Airports Company of South Africa-owned facilities alone	World-class aviation infrastructure provider, strongly driven by the need to meet legislated requirements. Delays and inconvenience due to continuous expansion to meet growth exceeding 10% per annum. A profitable company, with ample funding.	Potential need for increase in quantity due to continued high growth rates.
Ports	C+ for Transnet-owned facilities only	Useful life extended by proper management practices applied to ageing infrastructure. Increased investment and support under way to address increased demand. Further improvement expected as Transnet profitability improves.	Increase in quantity (new ports, and extension of existing port facilities). Performance improvements necessary to recover from backlogs experienced in some ports.
Railways	B for heavy-haul freight lines	World-class and well-maintained profitable iron ore and coal lines. Upgrading programmed where demand is approaching capacity.	Absence of any infrastructural backlog. Planning to increase capacity under way.
	C for general freight lines retained	Declining condition in recent years due to maintenance backlogs and skills reduction. Increasing traffic volumes, requiring urgent upgrading. Improvement expected as Transnet profitability improves.	Improvements in condition and performance urgently required.
	E for uneconomical general freight lines	Low-volume, low-priority lines in the process of being disposed of.	Reduction of quantity.
	D+ for passenger lines	Gradual deterioration due to inadequate maintenance funding, decreasing skills base, and vandalism, with resulting increased safety risks. Refurbishment under way. Improvement expected with the transfer to Department of Transport.	Improvements in condition and performance urgently required. Investments in quantity, in terms of projects such as the Gautrain, may be pursued in specific cases.
Electricity and distribution	C+ for Eskom's generating and bulk transmission capacity	Demand has nearly reached the limit of generating capacity. Shortfall due to be further depleted, before improving around 2011, when new base-load stations commissioned. Eskom profitable, with no shortage of funding, but capital programme delayed too long. Long and vulnerable transmission lines from Mpumalanga coalfields to urban centres. Risk of power cuts until the reserve increases.	Urgent increase in quantity of electricity generation required. Improvements in condition of transmission lines also required.
	C+ for Eskom's local distribution networks	Major and ongoing strides in provision of electricity since 1994 (also applying to municipal distribution networks). State of (Eskom local-distribution) infrastructure generally acceptable, but there are skills shortages.	Increase in distribution quantity, especially in rural areas (with the help of new technology). Improvements in condition of transmission lines.
	C− for municipal distribution networks in major urban areas	Inadequate operation and maintenance capacity, and shortage of skilled personnel. In many areas, ageing and/or overloaded infrastructure. Improvements discernible.	Improvements in condition of transmission lines. Skills required for maintenance operations.
	D− for municipal distribution networks in all other areas	Same types of problems as in major urban areas, but significantly worse.
Source: SAICE (2006) (amended).

Such evidence is supported by the Development Bank of Southern Africa (DBSA) Infrastructure Barometer Report of 2006. Four common constraints are identified at the end of the discussion on the current state of infrastructure in South Africa; namely, (1) institutional and capacity constraints on delivery, (2) lack of sufficient capacity to maintain infrastructure assets, (3) lack of regulation and efficiency, and (4) environmental damage (DBSA, 2006:136–8). The first constraint refers to the inadequate performance of the existing stock of infrastructure, and the second to its condition. These two common constraints featured in the DBSA report highlight the need to improve the quality of South Africa's current infrastructure.

3.1.1 Transport infrastructure

Comparative analysis of South African infrastructure measures with those of other countries and regions reveals that the quality of road infrastructure in this country lags behind that of other regions (Estache & Goicoechea, 2004). The measure of paved roads as a percentage of the total number of roads in South Africa (21 per cent) is significantly less than the world average (50 per cent) and that of all other income groups, although Bogetic and Fedderke (2006) argue that such a figure might be too extreme, and Perkins (2003) calculates that 31 per cent of South Africa's roads are paved. However, although the Perkins indicator is higher than that of Estache and Goicoechea, South African road quality is still seen as being far below the world average, with the usefulness of such a comparison being lessened somewhat by the fact that South Africa is an arid country, marked by large distances and low volumes of traffic. Commute times in the major cities (measured in terms of the number of minutes taken for a one-way work trip) further suggest the poor performance of the country's urban road network, although such timing also depends on the country's historical development patterns. South Africa (at 35 minutes) performs very poorly in comparison with all other income groups (even more so than do the low-income countries, at a comparative measure of 33 minutes) and the world average, which stands at 31 minutes.

Corroborating the findings of the Estache and Goicoechea (2004) data, the SAICE report suggests that the road infrastructure in South Africa requires serious attention. Although the coverage of the road network is good, the ‘most common cause of the failure of paved road surfaces is neglect – neglect of routine maintenance, and neglect to repair damage without delay’ (SAICE, 2006:10). The DBSA report notes that provincial and local roads in particular are seen as deteriorating: ‘It is clear that the quality of provincial roads has declined over the past decade. Road maintenance … is hampered by a lack of necessary skills’ (2006:137). The poor condition of the roads concomitantly increases the transport costs involved. The evidence suggests that logistics costs in South Africa make up 15 per cent of its Gross Domestic Product, while the country's trading partners' corresponding costs make up only 8.5 per cent of their Gross Domestic Product on average (Kularatne, 2006). Apart from the poor condition of the roads, the DBSA report suggests that the institutional arrangements entailed in managing the road infrastructure are complex, which results in local governments and agents shirking their maintenance responsibilities (DBSA, 2006:137).

The SAICE report also investigates the performance of airport and port infrastructure in South Africa. The high rate of national economic growth can clearly be seen to be creating a very high growth in demand for such infrastructure services. In general, the operation and management of the port and airport infrastructure is adequate – although, due to the growth in demand, it is the financial state of Transnet and its business units that underlies the infrastructural deterioration and obsolescence of South African ports (SAICE, 2006:12).

Given the earlier discussion of the impact that infrastructure is expected to have on trade, and the poor performance shown by the indicators of infrastructural quality in South Africa (Table 2), one would expect poor South African export performance. According to Edwards and Alves (2006), this is indeed the case. They argue that South African exports are not hampered by foreign export demand, but rather by supply constraints, especially lack of sufficient investment in transport infrastructure (Edwards & Alves, 2006:43). Even when considering the impact of infrastructure quality only on trade, the increasing importance of accelerated logistics for ensuring contracts for local manufacturers in time-constrained international markets makes a strong case for improving transport infrastructure quality (both its condition and its performance) in South Africa.

3.1.2 Electricity

According to the data collected by Estache and Goicoechea (2004), electricity transmission and distribution losses in South Africa do not seem particularly bad in relation to comparable areas and income groups in other parts of the world. However, the data were last updated in 2002, before the electricity supply shortages South Africa has experienced since 2005. One reason for such supply shortages was the apparent failure to maintain the transmission lines in the Western Cape, although Eskom (2006) blamed the environmental factors, such as the increased number of wildfires, and mist.6

The SAICE report, however, presents a different picture to that presented by Estache and Goicoechea (2004). Apart from the critical demand for increased generation capacity (quantity increases), the poorly maintained transmission lines (both the bulk grid lines and the municipal distribution lines) urgently require intervention. Such evidence is corroborated by that presented in the DBSA report, which notes that ‘expenditure on maintenance has been inadequate, and has resulted in increased numbers of failures and brownouts’ (DBSA, 2006:137). The lack of the requisite technical skills for undertaking such maintenance programmes, however, seems to be a serious constraint for Eskom.

3.1.3 Water and sanitation

While the need to increase the country's bulk water infrastructure capacity is evident, as is the need to improve distribution, especially to the rural areas, the poor condition of the distribution network is also highlighted in the SAICE report. The ‘poor and often unacceptable quality services’ are, according to the report, due to a lack of adequate skills at the municipal level, where high levels of leakage occur (SAICE, 2006:9). According to the DBSA report, the poor condition of the services is due to ‘the widespread lack of institutional capacity, both technical and financial, to operate and maintain water services and to generate sufficient revenue’ (DBSA, 2006:137).

The sanitation infrastructure presents an even more pressing predicament:

A nationwide sanitation sustainability audit in 2004/2005 to ascertain the functionality of sanitation projects completed since 1994 revealed that 28 per cent of households' sanitation facilities have failed or are in the process of failing, and only 53 per cent of municipalities have adequate operations and maintenance capacity. (SAICE, 2006:9)

shortage of relevant technical skills, especially at the local level, seems to limit the chance of any improvement to this sector.

3.1.4 Telecommunications

Telephone faults per 100 mainlines is the only existing measure of the quality of telecommunications infrastructure, although such a measure obviously does not assess the quality of all telecom infrastructure, such as, for example, the speed of Internet services. South Africa (at 48 faults) is significantly worse than the world average (at 37 faults) and the average of middle-income countries (at 25 faults). While the SAICE report did not assess any telecommunications infrastructure, the DBSA report argues that, spurred by increased competition from the cellular industry, the information and communications technology (ICT) sector is not allowing its infrastructure to deteriorate. ‘Almost by definition, ICT infrastructure is continuously renewed’ (DBSA, 2006:138). Therefore, while problems of access and high costs persist (due to Telkom's monopoly of the service), with a consequent impact on performance, the condition of the physical infrastructure remains of a high standard.

3.2 Addressing the issue of infrastructure

The poor condition and inadequate performance of infrastructure could be improved by simply building more of it. New infrastructure, although definitely required, should be built in order to satisfy demand; for example, for electricity generation and airport capacity – two areas where, although the condition and performance of the existing infrastructure is good, demand currently outstrips supply. However, the provision of new infrastructure cannot supplant the need for an improvement in the quality of the existing infrastructure. The construction of more ports (such as the deep-sea port at Coega in the Eastern Cape) will alleviate some of the burden of container traffic at other South African ports, but only if the supporting transport infrastructure (both road and rail) is in good condition and marked by reliable performance.

However, adding more infrastructure is not the only way to address the problem – in fact, it is probably one of the least efficient ways to use those resources that are currently available. Addressing key constraints to the logistics network, eliminating the red tape involved in extending the infrastructure, speeding up services, stabilising prices and increasing the number of private initiatives undertaken7 may serve as important alternatives to a simple cash-spending fiscal exercise. However, such measures are not easy to take – and are rather more difficult than just putting out a tender and signing a contract for a new project. Such measures require local-level managerial capacity, which is evidently lacking in areas where the effects of the constraints are the most severe.

Furthermore, politicians and bureaucrats involved in decision-making may lack the political will to approve projects for improving infrastructure quality (Fourie, 2006). For example, Robinson and Torvik (2005) point out that a new bridge or port facility attracts much more attention than does the filling in of potholes or the reduction of red tape, even if these smaller improvements lead to greater efficiency. They propose a theory of ‘white elephants’ (investment projects with a negative social surplus) – a political economy theory of inefficient redistribution that is politically attractive when politicians find it difficult to make credible promises to supporters. They show that such projects are politically appealing, due to their inefficiency:

This is so because it allows only some politicians to credibly promise to build them and thus enter into credible redistribution. The fact that not all politicians can credibly undertake such projects gives those who can a strategic advantage. Socially efficient projects do not have this feature since all politicians can commit to build them and they thus have a symmetric effect on political outcomes. (Robinson & Torvik, 2005:201)

Bureaucrats should not be incentivised in terms of the amount of resources spent on infrastructure provision but rather in terms of infrastructure performance criteria, such as higher turnover at ports and shorter waiting times for rail services.

4. CONCLUSION

Infrastructure investment, which is a serious constraint in South Africa, is being addressed within the government's AsgiSA framework. Yet the current empirical analysis of the impact of the infrastructure on growth still does not consider the impact of infrastructure quality, broadly defined. South Africa is lagging behind in terms of infrastructure quality measures. Poor infrastructure quality is increasing trading costs by restraining exports, a constraint that may even worsen South African export performance in the foreseeable future. Merely stating that more investment is required is inadequate. In order to be able to direct attention to the key constraints, we need to be able to describe them clearly. To do this, better measures of infrastructure quality, applied over a longer time span, are required. Such evidence could then be used to make the proper adjustments. Yet politicians and bureaucrats do not always follow the script. Policy-makers need to redirect their attention to improving the condition and performance of infrastructure – for example, improving port efficiency and timeliness, railway security, Internet speed and the physical road quality – rather than thinking that more is necessarily better.

Notes

¹The quality measures are, however, not as robust as the quantity measures. Calderón and Servén (2004) argue that such lack of robustness may reflect the poor measures of infrastructure quality, or the fact that the quality and quantity of infrastructure are strongly correlated.

²A gravity model predicts bilateral trade flows based on the economic sizes and distances between two countries. Variables are added to the basic equation to test whether other factors have a significant effect on trade between two countries.

³Although fast delivery is usually only relevant for the perishables market, Nordås and Piermartini (2004) note that time is also relevant in the clothing industry, particularly in the segments characterised by rapidly changing styles. Given that clothing is a labour-intensive industry but that most labour-intensive countries have poor infrastructure, countries closer to major world markets may benefit at the expense of the distantly located, but cheaper, exporters (as found by Evans & Harrigan, 2005).

⁴At best, proxies can be used to measure the quality of infrastructure. Kilometres of paved road is one variable that can be transformed to measure some type of quality, when divided by the total kilometres of paved roads. However, the series was discontinued by Statistics South Africa in 1994, which rendered any later analysis impossible.

⁵A more comprehensive investigation into South Africa's cross-country infrastructure performance has been undertaken by Bogetic and Fedderke (2006).

⁶Although the announcement was received with some scepticism by the public, an environmental impact on infrastructure is not uncommon. Diamond (2003:406) notes the serious degradation of Australia's infrastructure network due to climate change, pollution and natural disasters.

⁷The scope for public–private partnerships has increased since the 1990s, especially with the advent of new technologies that reduce the non-rival, non-excludable characteristics of pure public goods – although Kessides (2004), in an overview of the field, warns that such public–private initiatives should not be undertaken without the necessary reforms.