Infrastructure Development and Economic Growth in South Africa: A Review of the Accumulated Evidence

Abstract

This paper provides a broad overview of the relationship between infrastructure and growth, focusing on the South African case. The paper develops an intuitive theoretical framework in which to analyse this relationship, identifying five specific channels through which infrastructure may affect growth: as a factor of production, a complement to other factors of production, a stimulus to factor accumulation, a stimulus to aggregate demand and a tool of industrial policy. A framework is developed for evaluating empirical analyses of this relationship, which explores the implications of different definitions and measures of infrastructure and of potential data and estimation challenges. The empirical literature on South Africa is then assessed against this framework.

Introduction

The relationship between infrastructure and economic growth has, in recent years, become one of the most important economic topics in both academic and policy circles. The Accelerated and Shared Growth Initiative for South Africa (ASGI-SA) has identified inadequate infrastructure as one of the six most important constraints to growth in South Africa. The National Treasury allocated R416 billion to spending on infrastructure development and maintenance, broadly defined, in the current three-year budget cycle (National Treasury, 2007). This follows a period from 1976 to 2002 when annual infrastructure investment fell from 8.1% to 2.6% of GDP, with per capita expenditure falling from R1,268 to R356 (Fedderke and Bogetic, 2006a).

At the same time, academic journals have seen a flurry of infrastructure-related publications, beginning to correct a historical paucity of South African empirical research into the growth–infrastructure relationship. This research has followed the international trend of deploying increasingly advanced and more appropriate statistical techniques. Early international studies by Aschauer (1989a, 1989b, 1989c) and Munnell (1990) found a strong positive relationship between infrastructure and growth, sparking considerable academic interest. However, their findings have been widely criticised as relying on inappropriate techniques (Gramlich, 1994), and more attention is now paid to more recent studies that use more appropriate statistical methodology, such as Calderón and Servén (2004) and Estache, Speciale and Veredas (2005).

This paper provides a broad overview of the relationship between infrastructure and growth, focusing on the South African case. Section 2 briefly explores questions around the definition of infrastructure. Section 3 introduces the theoretical relationship between infrastructure and economic growth. A basic model of the determinants of economic growth is developed and used to explore five potential channels of effect from infrastructure to growth. Section 4 develops a framework within which to evaluate empirical research, presenting the available indicators, discussing the challenges facing the applied researcher and exploring some means of overcoming these challenges. Using this framework, the empirical evidence from South African studies is reviewed in Section 5. Section 6 explores the policy implications of this evidence, with particular emphasis on the funding needs and implications for infrastructure projects. Section 7 concludes.

Defining Infrastructure

A formal and detailed definition of infrastructure is not necessary for the purposes of this paper (for a discussion of such definitions, see Fourie, 2006). However, it is important to develop an intuitive understanding of the characteristics and types of infrastructure. This section develops such an understanding while introducing measurement issues that are central to Sections 4 and 5.

Infrastructure spending was historically defined as consumption expenditure by either government or the private sector but is now near-universally defined as capital expenditure, as infrastructure has been recognised as a capital good (Gramlich, 1994). Furthermore, infrastructure often possesses some characteristics of a public good: the owner or developer may struggle to exclude others from using it (non-excludability), and the benefits accruing to the economy as a whole typically exceed those accruing directly to the owner and even to the users (positive externalities) (Hirschman, 1958). These are not, however, necessary characteristics of infrastructure and some specific infrastructure items may be pure private goods.

In economic terms, infrastructure may legitimately be examined as a “stock” or a “flow” variable. In the former case, attention focuses on the stock of infrastructure at a given point in time. In the latter case, attention focuses on net infrastructure creation or loss over a given time period. Both infrastructure stocks and aggregate output (typically measured by GDP) are cumulative measures and thus possess marked time trends. Infrastructure flows and economic growth, however, are non-cumulative measures and are unlikely to possess such time trends. Infrastructure stock is thus generally compared to GDP and infrastructure flows to GDP growth, as differing time trends may otherwise obscure genuine relationships in the data.

By convention, infrastructure is broadly divided into two categories: economic and social. The former conventionally includes transport, communications, power generation, water supply and sanitation facilities, while the latter includes educational and health-care facilities, though some authors include cultural and recreational facilities (DBSA, 1998). This classification is largely ad hoc, as many forms of infrastructure may be considered as either economic or social. Educational facilities, for example, are widely defined as social infrastructure, but play an important role in generating human capital, which is also an economic function and carries important growth implications.

The two major approaches to measuring infrastructure are physical and financial. Financial measures calculate the depreciated value of the accumulated investment in a particular piece of infrastructure such as a road, school or power grid. Physical measures vary across different infrastructure measures: total length of paved roads, number of classrooms or total number of containers processed by a port. Constructing a single index of the physical characteristics of widely varying types of infrastructure is difficult, so it is common practice to use physical measures only when examining specific types of infrastructure. When examining aggregate infrastructure stocks or flows, it is more common to use financial measures. Some studies, however, use financial measures for infrastructure data disaggregated by type (Gramlich, 1994).

In the theoretical discussion that follows in Section 3, measurement issues are abstracted away. In Section 4, however, problems of measurement are explicitly addressed, as are some of the responses available to the empirical researcher.

A Basic Theoretical Model

Economic theory identifies five channels through which infrastructure can positively impact economic growth. To contextualise these channels, a brief diversion into basic growth theory is necessary. Economists typically describe growth in terms of a production function for goods and services, where aggregate economic output is a function of a collection of production inputs or “factors of production.” Models differ with respect to which factors of production they regard as the key determinants of aggregate output. Physical capital (machinery, equipment, etc.) is present in almost all models, while the size of the labour force, the level of human capital (the skills level of the workforce) and technology also make frequent appearances. Some models also include government policy, geographic features and institutions (such as judicial independence, protection of property rights and bureaucratic efficiency) as determinants of growth.

Most factors of production are positive determinants of aggregate output. Increases in the stock of physical capital, for example, are typically associated with increases in aggregate output. Thus, economic growth occurs when more factors of production become available and are put to use. However, it is also possible that some factors may exert a negative impact on aggregate output. Political instability, for example, has been shown to negatively affect aggregate output in South Africa (Fedderke et al., 2001). Furthermore, decreasing marginal product arguments apply here directly.

Discussion to date has focused on “supply-side” accounts of economic growth, in which aggregate output is modeled as the outcome of a production process. Some alternative models are “demand-side” or consider both demand and supply sides of the economy. These models regard aggregate output as the result of interacting supply and demand factors — implicitly assuming that output will only be produced if consumers (domestic or international) wish to purchase it.

Demand-side explanations of economic growth have experienced periods of popularity but have become increasingly uncommon in recent years. A relative consensus now exists among economists that demand-side considerations are important in determining short-run fluctuations in economic performance but play little role in determining medium- and long-run patterns of economic growth (Romer, 2001). In line with this consensus, this paper focuses primarily on supply-side linkages between infrastructure and growth.

Having developed a basic analytical framework, discussion now proceeds to the five potential linkages between infrastructure and economic growth.

Infrastructure as a Factor of Production

Infrastructure may be regarded as a direct input into the production process. As infrastructure could otherwise be placed under the broader heading of physical capital, this approach assumes that infrastructure may be related to growth in a manner different to other forms of physical capital (Gramlich, 1994). Theory holds that an increase in the stock of infrastructure would increase the output of the economy as a whole, directly inducing economic growth.

The role of power generation infrastructure provides a concrete example. It is a necessary input into many production processes for both goods and services, and unreliable power supplies render these processes either more expensive or impossible. This renders what economists call “marginal” transactions unprofitable  transactions that were previously just profitable become unprofitable. The net result is a decrease in the sum total of economic activity.

Infrastructure as a Complement to Other Factors

Alternatively, infrastructure may be regarded as a complement to other inputs into the production process, in two senses. First, improvements in infrastructure may lower the cost of production. Inadequate infrastructure creates costs for firms, who may have to develop contingency plans against infrastructure failure or even build infrastructure themselves. Inadequate transport infrastructure, for example, incurs potentially massive costs for firms who must find alternative means of transporting inputs and finished goods. The total cost of firms providing transport infrastructure may be far higher than the equivalent cost of state provision, as the latter can achieve economies of scale. A similar analysis applies to power generation infrastructure.

Good infrastructure generally raises the productivity of other inputs. The productivity of capital such as machinery is raised by reliable power supplies, while the productivity of labour will be higher if good education and healthcare infrastructure produce a well-educated and healthy workforce. “Total factor productivity” (TFP) is output growth over and above that attributable to the accumulation of factors of production — roughly, the efficiency with which factors are combined. Infrastructure may influence TFP by providing transport facilities necessary to operate decentralised production processes. These effects allow greater output for a given level of input and lower the cost of that input.

At the extreme, inadequate infrastructure may render some production processes nearly impossible. International trade, for example, depends on sophisticated transport and communications infrastructure. Modern trade theory and empirical evidence emphasise the importance of linking into cross-border production networks. Industrial agglomeration — the tendency for new industries to locate near existing industrial concentrations — also depends critically on adequate infrastructure (Mayer, 2003; Redding & Venables, 2004). The success of the motor industry in the Eastern Cape, for instance, would have been difficult without nearby harbours, road networks and reliable power.

A related example is tourism: provision of tourist services to international markets depends on transport infrastructure, and local provision is constrained without such infrastructure (Fourie, 2006).

Infrastructure as a Stimulus to Factor Accumulation

Each factor of production (e.g., human capital) is itself the outcome of a production process. Infrastructure — in the form of schools, roads used to access schools and electricity provided to schools — is likely to be an important input in human capital formation. More generally, infrastructure is a determinant of many factors of production. Thus, infrastructure may influence growth indirectly by boosting accumulation of other factors or by enhancing their productivity. These first three channels are captured graphically in related literature.

Infrastructure as a Stimulus to Aggregate Demand

Large infrastructure projects typically involve significant expenditure during construction and maintenance, increasing aggregate demand. Governments have often used large-scale projects as stimulus policies during recessions. Critics argue this effect is short-lived: without increases in productive capacity, extra demand merely raises imports and inflation. Mariotti (2002) finds that large-scale government expenditure in South Africa provides only a temporary stimulus to output (approximately two years) but produces long-term inflation. While infrastructure spending may increase productive capacity, its primary long-run impact on growth is more likely through the supply-side channels discussed above.

Infrastructure as a Tool of Industrial Policy

Government may invest in specific infrastructure projects to guide private-sector investment decisions. A road in a rural area may be intended to integrate the area into the regional economy and promote private investment. This approach has underpinned projects such as the Maputo Corridor and the Coega Development Corporation. Critics counter that such interventions often fail to stimulate growth, or simply divert resources from more productive uses.

Despite disagreement about the relative importance of channels, there is broad agreement that infrastructure and growth are positively related. Specific projects, however, can have negative impacts, especially where over-provision occurs (Canning & Pedroni, 2004). Empirical evidence on this remains ambiguous due to methodological and measurement challenges.

Growth as a Determinant of Infrastructure Spending

Growth may itself drive infrastructure investment: higher output generates demand for more infrastructure (transport, communication, training). Private firms might invest directly or lobby government for infrastructure. This “reverse causality” poses a significant challenge to empirical researchers (discussed in Section 4).

A Framework for Empirical Analysis

Empirical testing of the theoretical linkages in Section 3 poses significant challenges. This section explores available indicators of the infrastructure–growth relationship and discusses their merits and limitations. It then outlines data and estimation challenges and potential empirical tools for overcoming them.

Indicators

The first indicator of the growth impact of infrastructure is direct: the correlation between infrastructure stock and aggregate output, or between infrastructure investment and economic growth. This may be a simple correlation or a partial correlation that holds constant other factors. Researchers might, for example, calculate the correlation between infrastructure stock and aggregate output while controlling for the general level of physical capital. This avoids biased measures of the infrastructure–growth relationship that could be confounded by correlations between physical capital and growth (and other omitted variables.

Assessment brief

• Purpose: Critically analyse the relationship between infrastructure and economic growth (South African focus) using the supplied paper and related literature; produce a coherent, evidence-based assessment that follows academic conventions.

• Core tasks:

  • Summarise the theoretical framework (five channels: infrastructure as factor of production; complement to other factors; stimulus to factor accumulation; demand stimulus; industrial policy).

  • Explain measurement issues (stock vs flow; physical vs financial indicators) and empirical challenges (reverse causality, omitted variables, data limitations).

  • Review and evaluate empirical evidence for South Africa against the paper’s framework.

  • Discuss policy implications (funding needs, project selection, efficiency) and conclude with balanced recommendations.

• Deliverables expected: concise assessment brief + structured report that includes introduction, theory, empirical framework, evidence review, policy implications, conclusion, plus references to the key readings (e.g., Fedderke & Garlick; Gramlich; Aschauer; Munnell; Calderón & Servén; Estache et al.).

• Format & standards: academic style, accurate citations (APA/Harvard), clear signposting of assumptions, and demonstration of critical thinking rather than mere summary.

How the Academic Mentor guided the student

1. Clarify the brief & scope

   • Mentor reviewed the assignment requirements with the student, defined the boundaries (South Africa focus; time period; empirical vs theoretical balance), and set two measurable objectives: (a) explain the five channels clearly, (b) appraise South African empirical studies using the framework.

2. Identify and prioritise sources

   • Mentor directed the student to the core paper (Fedderke & Garlick) and to complementary studies (Aschauer; Munnell; Gramlich; Calderón & Servén; Estache et al.).

   • Advised to locate empirical South African studies cited in the paper and recent critiques to address methodological concerns.

3. Structure the report

   • Mentor helped produce an outline: Abstract → Introduction → Definitions/measurement → Theoretical model (five channels) → Empirical framework → Evidence review → Policy implications → Conclusion → References.

   • Emphasised signposting sentences at the start of each section for reader clarity.

4. Research & note-taking

   • Mentor taught focused reading techniques: extract thesis, methods, data sources, main findings, and limitations from each article.

   • Student maintained a working bibliography and annotated notes for each source.

5. Develop the theoretical section

   • Mentor coached the student to explain each channel with a concrete South African example (e.g., power outages and firm productivity; transport links and trade/tourism).

6. Empirical critique

   • Mentor modelled how to evaluate empirical papers: check definitions (stock vs flow), measurement, identification strategy (controls, instruments), potential reverse causality, and robustness checks.

7. Drafting & feedback cycles

   • Student produced drafts; mentor provided iterative feedback (clarity, argument flow, citation accuracy). Emphasis placed on avoiding descriptive summary only  every claim tied to evidence.

8. Polishing & referencing

   • Mentor reviewed referencing style (APA/Harvard), ensured all quoted facts had sources, and checked that limitations and alternative interpretations were acknowledged.

Final outcome

What was produced

• A structured assessment/report that:

  • Clearly summarises the five theoretical channels and measurement issues.

  • Critically evaluates South African empirical studies using the paper’s framework (highlighting strengths and methodological weaknesses).

  • Discusses policy implications (prioritisation of projects, funding instruments, monitoring of outcomes) and offers balanced recommendations.

  • Contains an annotated reference list (Fedderke & Garlick; Gramlich; Aschauer; Munnell; Calderón & Servén; Estache et al. and relevant South African studies).

Learning objectives met

• Demonstrated understanding of infrastructure definitions and measurement (stock vs flow; physical vs financial).

• Applied theoretical models to real-world South African examples, showing linkage between infrastructure and growth.

• Critically appraised empirical methods and evidence — including recognition of reverse causality and data challenges.

• Developed policy-relevant conclusions grounded in evidence.

• Practised academic research skills: targeted literature review, synthesis, structured writing, and correct referencing.

Final note from the mentor

• The student now has a publishable-quality assessment framework and a clear roadmap for follow-up empirical work (e.g., selecting a particular infrastructure type, constructing appropriate indicators, or exploring identification strategies to test causality). If you want, I can turn this into a full 1,500–2,000 word student report or prepare an annotated bibliography and a sample reference list in APA or Harvard. Which would you like next?

Need Help Understanding This Assignment? Get Safe, High-Quality Support Here

Before you wrap up, make sure you get the most out of the resources available on this page. You can download the sample solution provided, which is designed to help you understand the structure, formatting, and academic approach required for similar assignments. This file is strictly for reference and learning purposes only. Submitting it as your own work may lead to plagiarism penalties under your institution’s academic integrity rules.

If you want a submission-ready solution, our team of professional academic writers can craft a 100% original, plagiarism-free assignment tailored to your topic, university guidelines, and deadline. This ensures you receive high-quality content that is safe to submit and built around your specific requirements.

Why Order a Fresh Custom Solution?

• Fully original and plagiarism-free content

• Written by qualified subject-matter experts

• Proper academic formatting and referencing

• Delivered on time, even for urgent deadlines

• Helps boost grades while keeping your work compliant

Plagiarism Disclaimer

The downloadable sample is for reference, guidance, and study use only. Do not submit it as your own academic work. Always use original content when turning in an assignment.

[Download Sample Solution]    [Order Fresh Assignment]