Abstract
This article examines the design and implementation of the Expanded Public Works Programme, drawing from the lessons of the Gundo Lashu programme on labour-intensive construction in the road sector. The aim of the study was to examine major barriers to the design and implementation of the programme. The article is based on the analysis of data that were gathered from the Gundo Lashu programme. The data were collected using a structured questionnaire, face-to-face interviews with the project participants as well as reviewing key secondary data from the project documents. The study found that the projects under the Gundo Lashu programme have clearly demonstrated the importance of labour-intensive approaches, especially in the road sector. However, these projects were constrained by a wide range of challenges, ranging from a lack of community participation to poor product quality and inadequate planning for sustainable financing.
1. Introduction
The Expanded Public Works Programme (EPWP) was initiated in South Africa to draw significant numbers of the unemployed into productive work (Phillips, Citation2004; Department of Public Works, Citation2007). Key programmes within the EPWP were initiated in 2004 focusing on four key sectors, namely; the infrastructure sector, which is focused on increasing the labour intensity of government-funded infrastructure projects; the environmental sector, focusing on creating work opportunities in the environmental programmes; the social sector, targeting the creation of work opportunities in public social programmes; and the economic sector, which is focused on the development of small businesses and cooperatives.
The Gundo Lashu programme was initiated in 2001 in the Limpopo province as a pilot programme. The primary objective of the Gundo Lashu programme was to provide good-quality and cost-effective rural roads using labour-intensive construction methods. The programme was targeted to generate one million workdays of employment for the participants. Between 2001 and 2009 the Gundo Lashu programme implemented a number of road construction projects using labour-intensive methods as the dominant method of construction. The projects that were sampled for the study are located in the Vhembe District, under the Makhado and Thulamela local municipalities (RAL, Citation2009). presents a list of the sampled roads constructed and maintained using labour-intensive methods.
Table 1: Sampled projects
The main objective of the study was to examine the constraints and challenges experienced in the implementation of labour-intensive methods in the Gundo Lashu programme. The article focuses specifically on examining various barriers and constraints encountered both at the design and implementation stages of each sampled project.
Harmse (Citation2004) reflects that the research methodology used in both economic and transport geography has changed over time, partly due to the social nature of science and the struggle between competing philosophical positions. In this case, the article is based on analysis generated from the structured questionnaire and personal in-depth interviews. The study sample covered a total number of 128 respondents from the total population of 1339 people who participated in the Gundo Lashu labour-intensive projects. The respondents were identified through an employment database that was provided by the Gundo Lashu project contractors.
The data collection instruments including the structured questionnaire and personal in-depth interviews were conducted both with the Gundo Lashu project participants and with non-participants who were sampled from the project and non-project communes. Secondary data sources used included the review of various literature on targeting, administrative data from the Gundo Lashu programme and various policies and legislations on the EPWP and labour-intensive construction.
Quantitative data analysis was carried out using Stata version 10. Comparisons were made between the two groups (control and treatment) based on odds ratios estimated from conditional logistic regression analysis. Thematic analysis was also used to analyse qualitative data collected through personal interviews. In this case, descriptive data and evidence relating to each question were classified into distinctive classes based on their common qualitative characteristics. Various opinions expressed by the respondents were grouped in order to determine the priority structures to analyse both conflict and complementary responses which were expressed differently by the respondents.
Historically, public works programmes were initiated to address various socio-economic development challenges. A government labour-based public works project appraisal report (African Development Fund, Citation2007) reflects that public works programmes were generally developed to serve as relief programmes for emergencies created by catastrophes. Within the transport sector, the key objective was to provide income, assets, employment and food security through the repairs of damaged roads and bridges. Von Braun et al. (Citation1992) reflect that countries such as Kenya have long-standing experience with public works programme while others such as Botswana, Ethiopia and Zimbabwe have had experience with public works programme designed specifically to achieve the objective of food security during emergencies. According to Von Braun et al. (Citation1992), since the 1960s countries such as Mali and Senegal have also experienced public works programme which were developed either as drought relief programmes or as International Labour Organisation pilot projects. In contrast, public work programmes in South Africa were developed to address key developmental challenges.
The South African Rural Transport Strategy classifies these challenges under three categories; first, the historical backlogs in road infrastructure development with the continuing under-investment in road infrastructure and reactive responses to the road maintenance programmes in rural areas; second, the sheer physical remoteness and low population densities of some areas, the attendant transport provision, accessibility, vast networks of poor quality, low-volume roads and low demand thresholds; and finally, the existing technical and managerial weaknesses of most rural municipalities, which in turn present formidable capacity-building requirements (Department of Transport, Citation2003:11).
Targeting of public works programmes plays an important role in determining the success rate of the programme. Targeting seeks to deliver benefits to a selected group of participants, in particular poor and vulnerable people. Targeting mechanisms attempt to link a project's specific purposes with its intended group of beneficiaries. Gelbach & Pritchett emphasised that there are generally three conditions that must be satisfied to ensure proper targeting for both the programme locality and to ensure that the right people receive the benefits: firstly, actionable reasons for poverty must be identified; secondly, programmes must be devised that target these particular reasons; and lastly, efforts must be made to direct these programmes towards the people who most need them.
The subsequent section elaborates on various targeting approaches that are more relevant to the Gundo Lashu programme in order to assess their impact thereof on success and limitations of the programme. The second part of this document explores barriers which impacted on the Gundo Lashu programme and demonstrates how these barriers influenced the programme design, implementation and outcomes.
2. Major findings
2.1 Summary of programme impacts
The Gundo Lashu programme used labour-intensive methods as the dominant method of construction. Analysis of the programme's impacts showed that the Gundo Lashu programme yielded mixed effects. Most notable is that fact that there was little evidence generated from the study to demonstrate that the road construction projects within the Gundo Lashu programme have had a direct impact in terms of reducing the poverty of the people living in the sampled study areas and the project participants. The programme achieved its economic development objectives in terms of key indicators of performance, which amongst others include the total road length constructed, total number of jobs created and so forth.
The Gundo Lashu programme constructed 222.8 km of roads by labour-intensive methods using cement and slabs, natural pavement, emulsion treated and composite emulsion treatment base. In line with this, a total of 6827 jobs and 1 285 189 person-days of work were created from the years 2004 to 2009. This is a significant contribution towards redressing the problem of unemployment. In terms of gender breakdown of these jobs, the programme created a total of 3206 jobs for women during the project lifecycle. A total number of 2996 jobs were created for youth below the age of 35 years. In essence, the Gundo Lashu programme had positively impacted on the quality of the labour supply based on the training and work experience provided to the participants during the duration of the project (Musekene, Citation2013).
The study also found that the programme provided basic income which was paid in the form of direct wages to those who participated in the projects. These incomes were important to supplement other household earnings from various sources. The study also found evidence of the development of very small and survivalist businesses by both the project participants and other local community members who reside in areas where labour-intensive road projects were initiated (Musekene, Citation2013).
According to Musekene (Citation2013) the impact of Gundo Lashu labour-intensive projects on employment yielded mixed results. It was, however, noted that the development of the Gundo Lashu projects created a reasonable number of jobs; but these jobs were mostly part-time, short term and unsustainable (Musekene, Citation2013). In this case, the study points to the fact that such a minimal contribution from the programme did not help reduce the incidence of unemployment. While a few projects have successfully targeted groups in extreme poverty, the outcomes of these programmes on unemployment were generally quite unflattering. This situation makes the Gundo Lashu programme rather a non-efficient tool in the ambition to reduce unemployment.
According to Musekene (Citation2013), other the social dimensions related to poverty reduction, sustainable livelihoods and opening of access to market opportunities in the Gundo Lashu programme were not sufficiently achieved. It was revealed that the programme did not independently have a significant impact on the communities who participated in the projects compared with those communities and individuals who did not participate in the projects. The project participants’ socio-economic status did not change significantly from those communities without projects. The programme's impact on access to both educational and health facilities was statistically insignificant or minimal at the 95% significance level, and hence the study concludes that the accessibility to new roads by the communities did not have a significant impact on the reduction of travelling time to both the educational and health facilities (Musekene, Citation2013).
2.2 Planning and implementation challenges for the Gundo Lashu programme
Generally, the planning for municipal infrastructure such as roads is the responsibility of the local government. The Municipal Structures Act 1998 (Act 117 of 1998) and the Municipal Systems Act 2000 (Act 32 of 2000) provide for municipalities to integrate and coordinate plans and programmes with other spheres of government and stakeholders. The Municipal Systems Act 2000 (Act 32 of 2000) outlines the process for greater community participation by encouraging municipalities to create conducive environments for job creation and the provision of basic services in an effective and efficient manner.
The analysis of the project administrative data and document reviews highlighted a number of key challenges and barriers which are critical to the success of the Gundo Lashu programme. Key among these barriers is the management planning and targeting barrier, human resource barrier, functional and structural barrier and financial resource barrier.
2.2.1 Management planning and targeting of the programme
The concept of planning and targeting has long evolved in the history of development theories, especially as it relates to poverty programmes. In its simplest form, the concept entails the selection of the suitable local areas for implementation, eligible people to receive benefits and the establishing mechanisms for identifying those individuals within the population. The rationale for targeting is that, especially when resources are limited, it is better to give a smaller group such as the poor a greater amount of support, rather than provide a smaller amount of support to everyone (Elbers et al., Citation2004).
McCord & Van Seventer (Citation2004) believe that the EPWP can be a part of building human and social capital, and could be even more significant if the programme targets even poorer people. The EPWP Code of Good Practice recommends the utilisation of local resources, specifically the recruitment of project participants from communities within the 7 km radius from the road corridor. The EPWP Code of Good Practice, however, did not prescribe to the contractors how labour should be recruited. For this reason, EPWP contractors in the study area worked closely with community structures, including the Community Liaison Officer and local civic structures, in the identification of potential project participants. The following section provides a summary of key lessons from the study.
The targeting process of various projects within the Gundo Lashu programme was not fully regulated. The study found that the decisions on the communities from which people should be recruited was left for the contractors to decide. Hence, the participant selection process was open to abuse. This led to some of the participants being recruited into different projects more than once, recruitment of participants who did not meet the programme requirements and the recruitment of people outside the targeted population.
The study found that the programme did not fully reach its intended beneficiaries. Communities targeted by the Gundo Lashu programme also had lower variability since respondents from the targeted communities did not demonstrate any positive benefits over untargeted communities. One of the key objectives of the Gundo Lashu programme was to increase the accessibility of communities to services such as health and education, and therefore include people in economic activities.
The results of the study on the impact of the projects on travel times and access to services such as educational and health facilities did not vary. The outcome of the statistical analysis reflected in – shows that although the mean travel time to both the educational and health facilities was less for the treatment group than the control group, the accessibility to roads did not have a statistical significant impact on travelling time to health and educational facilities.
Table 2: Variance ratio test on access to educational facilities
Table 3: Comparison of means on access to educational facilities
Table 4: Variance ratio test on access to health facilities
Table 5: Comparison of means on access to health facilities
Access to educational facilities, in terms of the time used to reach these facilities, was used to measure the impact of the programme on both the control and the treatment groups. However, before comparing the means of the two sampled groups, the variance of the two independent sample groups were compared using the variance ratio test (sdtest) and the results are presented in . Since p = 0.2439 > 0.05, it was concluded that there is similar access to educational facilities for both the treatment and the control communes.
The mean access to educational facilities was compared by making use of a two-sample t-test with equal variance. The results are presented in . The p-value is 0.1407 > 0.05, which implies that the mean access to educational facilities was equal at the 0.05% level of statistical significance for both the treatment and the control groups. The mean access to educational facilities for the treatment group was 32.34 minutes, with travel times ranging from 5 to 90 minutes. The control group, on the other hand, had a mean access to educational facilities of almost 37 minutes, with travel times ranging between 15 and 80 minutes. It was therefore concluded that although the mean travel time to educational facilities was less for the treatment group (32.34 minutes) than the control group (37 minutes), the accessibility to roads did not have a significant impact on travelling time to educational facilities. These results challenge the notion that rural transportation and access to roads in rural areas significantly increases the access of local communities to services.
The travel to health facilities was used to determine the impact of EPWP road infrastructure on travel time. The variance of two independent sample groups were compared using the variance ratio test (sdtest) and the results are presented in . Since p = 0.3156 > 0.05 we concluded that there was similar access to health facilities for both treatment and control communes.
The mean access to health facilities, using a two-sample t-test with equal variance, was compared and the results are presented in . The p-value was 0.2606 > 0.05, which implies that the mean access to health facilities of both the treatment and the control groups is equal at the 0.05% level of statistical significance. We therefore concluded that although the mean travel time to health facilities was less for the treatment group (36.01 minutes) than for the control group (39 minutes), the availability of roads did not have a significant impact on travelling time to health facilities.
According to (Van Domelen, Citation2007:9), the question of ‘who benefits from public spending is a deeply political question that goes to the nature of power relations within a society’. Various literature on the political economy of public spending, including De Donder & Hendriks (Citation1998) and Van de Walle (Citation1999), highlight many reasons why spending decisions are made not on the basis of technocratic criteria but on calculations of political power, patronage and favouritism. Power is thus consolidated by controlling the levers of public spending (Gelbach & Pritchett, Citation1999; Van Domelen, Citation2007). The study found that the targeting approach for the Gundo Lashu projects was not fully aligned with other government targeting approaches. The study revealed some major limitations on the Gundo Lashu targeting approaches in relation to targeting mechanisms used in other government sectors, especially the means-test targeting approach used in the government's social assistance programme. These inconsistencies have had a profound negative impact on government's efforts to develop and support a cohort of rural poor people who require government interventions.
Based on the analysis of the secondary data, it was evident that the prioritisation of projects within the Gundo Lashu programme should have taken advantage of all opportunities that pertain to the establishment of infrastructure in the broader public interest. This would have resulted in ownership of the programme by local communities. Furthermore, in pursuit of this approach, the prioritisation of projects within the Gundo Lashu programme would instil a desire to pursue investment where the programme is composed of projects aimed at creating infrastructure, geared particularly towards the needs of the poor. These findings are in line with the findings of McCutcheon (Citation1989:119), who pointed out that labour-intensive methods are widely regarded as backward and rejected as incompatible with the modern world. Even after construction by means of labour-intensive methods, the taint of backwardness still lingers.
In terms of the prioritisation of projects, the Spatial Planning Framework for South Africa emphasises that development projects should be created around the demands of the local communities and aligned to the existing infrastructure. The results of the study, however, show that unnecessary infrastructure projects within Gundo Lashu were created in various communities in order to ensure that EPWP projects take place. Interviews with Road Agency Limpopo (RAL) showed that the agency attempted to meet the integrated development plans objectives by adhering to the prescribed planning processes. Contrary to this, another secondary source showed that the planning process for most of the projects under the Gundo Lashu programme were ad hoc in nature. Some projects were initiated as a result of a crisis, some lacked spatial focus and projects often did not have any links to municipal or local rural development and infrastructural planning. An example is the Madabani project, which was initiated in response to a train and bus accident which killed 25 people in 2007.
Interviews with community members also revealed that local communities did not want roads constructed using labour-intensive methods in their areas and would prefer contractors to use conventional methods. This raised questions regarding the significance of employment generated through these projects and alleviation of poverty levels. It creates an impression that most communities do not regard labour-intensive road projects as a value-adding activity.
In addition, the study found that the poor condition of the roads constructed within the Gundo Lashu programme, in most study locations, were linked to institutional weaknesses of the local municipalities. A common assumption about labour-intensive work is that it is only pick-‘n'-shovel work because the tools and techniques are simple; the work involved is not ‘proper’ engineering and therefore does not require thorough analysis and planning. This attitude is, according to McCutcheon (Citation1989), exacerbated by the lack of appreciation of the sound theoretical basis for the substitution of labour for equipment, because it is not known that good results can be achieved and no attempt is made to be competitive. The information obtained from the participants in the Gundo Lashu programme also revealed that the poor standard of workmanship on some roads undoubtedly contributed to these negative attitudes towards labour-intensive construction methods.
With regard to the maintenance of projects after completion, the study found that there was a lack of a proper infrastructure maintenance plan. Long-term social and economic benefits from roads are often threatened by a neglect of periodic maintenance. Rural roads, particularly paved roads, quickly deteriorate if not regularly maintained and their benefits can be lost if they are periodically impassable or the overall condition is bad. This results in increases in vehicle maintenance costs and time.
2.2.2 Human resource barrier
Human resources – that is, people, including project participants and the project management team – are critical to the performance of the project. The development and implementation of road projects involve a huge investment in human resources both at the design and implementation stages. The RAL report reflects that during the project design stage, human resources – which is categorised as consultants, specifically civil consultants, project management and onsite supervisory staff and contractors – were significant for the effective implementation of projects under the Gundo Lashu programme. In line with the above, the inherent requirements for the programme were that the consultants and implementers should be trained in labour-intensive construction methods, should have experience in labour-intensive work and should demonstrate their commitment to the objectives of the programme.
The study found that limited project management skills compounded by inadequate knowledge of labour-intensive management and supervision hindered the performance of some projects. As a result, some projects were forced to switch from labour-intensive to capital-intensive in order to meet the set project target dates for completion.
Poor management capacities occur at the municipality and department levels. While there are a number of management barriers that exist within the programme, findings of the study revealed that districts and local municipalities have to a large extent not been able to realise their planned targets for the projects under the Gundo Lashu programme. This was a result of inadequacies both from the Department of Labour, district councils and the province. This, in particular, relates to lack of policy direction, but also incorrect application of the defined policy and guidelines. These findings reaffirm the conclusions of Mashiri et al. (Citation2005) that governance issues in the management of community-based projects are critical and that there is a need for a legitimate, balanced and representative institutional framework underpinned by a consistent management structure, which is appropriately located in the overall scheme of things. Within the Gundo Lashu programme, these shortcomings resulted in the municipalities not discharging their roles and responsibilities, such as supervision, monitoring and evaluation, as intended. A number of the inadequacies identified from the secondary data sources include the following:
Inappropriate decisions and recommendations related to the recruitment and appointment of project workers.
Inappropriate decisions related to the products used for road surfacing.
Undue and unwarranted vertical and horizontal reporting.
2.2.3 Financial resource barrier
McCutcheon (Citation2008:27) argued that previous experience related to employment creation through public infrastructure and building works in South Africa revealed that most of the projects were not planned in line with required financial resources. This resulted in the production of ill-defined products of doubtful value through processes which did not lead to the building of individual, community or institutional capacities. Phillips (Citation2004:8) outlined that the EPWP emerged with a different funding mechanism compared with previous initiatives. The Department of Public Works identified R15 billion of the Provincial and Municipal infrastructure grants for a five-year period (2004–09), to be spent on building infrastructure under the EPWP. The projected amount of infrastructure that the provinces and municipalities were to construct through the EPWP during this period included 1500 km of roads, 1500 km of storm water drains and 150 km of urban sidewalks.
At the project level, the study found that the existing funding mechanism and arrangements created a challenge regarding funding for future maintenance of the existing projects. An analysis of the municipal Medium Term Expenditure Framework budget, however, revealed a lack of funding dedicated to the Gundo Lashu programme. The lack of funding can potentially stall future maintenance of the infrastructure. The establishment of regular and transparent maintenance regimes and criteria for rehabilitation becomes difficult after the project completion, and as a result most roads were left without maintenance funding. Roads consequently were constructed, and due to lack of maintenance they deteriorated and then had to be reconstructed again. As a result, communities experience peaks and troughs of accessibility, rather than having a constant and guaranteed level of access.
The findings of the Amadiba road project study (Mashiri et al., Citation2005) revealed that the substantial impact on the community during the life of the project and long-term sustainability are engendered by a significant amount of project funding that needs to remain within the community, to feed the local economic circuits with a view to enhancing project multiplier effects. In contrast, the data showed that various planned programmes within the Gundo Lashu programme experienced inappropriate funding relative to the scope of the projects. The resource barrier which is mostly in terms of funding and the financial barrier resulted in a number of key projects planned for implementation remaining at a conceptual phase while others were deferred. This has impacted negatively on the delivery of services and the scope of reach for the programme in the district.
2.2.4 Functional and structural barriers
Identification of and prioritisation of labour-intensive road projects was made during the conceptualisation of the Gundo Lashu programme and agreements reached with the host and local municipalities. These roads were further linked to the municipal infrastructure development plans. Owing to the changing nature of the municipal infrastructure development plans, various projects identified during the initial stage were implemented in the later years; this resulted in a mismatch between the municipalities’ current infrastructure plans and the initial agreement with RAL.
It was conceived that there would be proper collaboration and communication between RAL and the host municipalities, but that was not the case. Lack of proper planning and implementation arrangement between RAL and municipalities resulted in a separation between these structures. These separations were as a result of unclear mandates and structural barriers that existed between RAL and the municipalities where the municipalities felt that the labour-intensive project was imposed on them. As a result, most municipalities were not involved in the planning and implementation process which resulted in inappropriate handover of the completed road infrastructure to local municipalities. The study revealed that the lack of involvement by most municipalities was caused by a lack of resources for future maintenance by municipalities, limited knowledge of labour-intensive construction projects by municipalities and structural challenges resulting from roads which fall within the cross-boundary municipalities.
The most common challenge within functional and structural barriers was communication and coordination. The study found that a major weakness of the Gundo Lashu programme is that of communication and coordination which is linked to political and strategic direction. The study found that there was no formal hand-over of the projects from the RAL to the respective municipalities during the project completion phase. As a result, roads which were constructed within the Gundo Lashu programme may not be properly maintained in the future, due to a lack of succession and integrated planning processes and synergy between the municipal local economic development plan and the Gundo Lashu programme. This can decrease the benefits expected to be derived from the project.
Undoubtedly improved communication and coordination vertically between national, province and municipal spheres of government and horizontally across districts and municipalities was identified as one of the major challenge in the successful implementation of the Gundo Lashu programme. This emanated from the fact that the responsible department does not have the necessary authority to insist on participation from other departments. In particular, it is widely recognised that the link between the implementing agency (RAL), the Department of Public Works, municipalities and the project communes is poor, and in some instances is non-existent.
Furthermore, projects under the Gundo Lashu programme were designed and approved with a monitoring and evaluation framework of limited scope. The existing framework for monitoring and evaluation does not focus on results but is driven from the traditional monitoring and evaluation system design which focuses on inputs and outputs, specifically in line with the emphasis on monitoring and reporting on budgets. As a result, very few of the Gundo Lashu programme monitoring and evaluation reports captured information on important lessons that could benefit future planning. The study identified various gaps within the reporting, monitoring and evaluation system for the Gundo Lashu programme, which included lack of a central knowledge management and archive system for the documentation of good practices and lack of peer-reviews for project reviews.
3. Conclusion and recommendations
This study identified various barriers which affected the Gundo Lashu programme both at the planning and the implementation stages. Key amongst those identified were the management and planning barrier, the human resource barrier, the financial resource barrier, and the functional and structural barrier. The study further concludes that the planning and implementation of Gundo Lashu programme was also highly influenced by external political motives with limited pre-feasibility studies, which has resulted in a number of challenges. Key amongst these challenges were lack of support and buy-in from local communities, projects which were abandoned prior completion, low quality of roads constructed, lack of proper plans and inappropriate budgets for future maintenance. Undoubtedly, lack of coordination and strategic leadership vertically between national, provincial and municipal spheres of government and the implementing agency, and horizontally across districts and municipalities, was one of the key challenges faced in the successful implementation of the Gundo Lashu programme.
There is therefore a need for a comprehensive master plan for the Gundo Lashu programme which is integrated with rural road network planning at local government level. This would be an ideal tool to increase the effectiveness of rural road investment and stakeholder involvement. The central task of the master planning process would be to identify, from a large number of rural roads in the project area, a core network that would provide the basic minimum road connectivity between villages and centres of economic activity, based on the existing road networks and the travel patterns of the local communities.
While acknowledging that the Gundo Lashu programme is not entirely poverty targeted, the study emphasised the significant contribution the programme had to addressing the problem of poverty. It is therefore recommended that further criteria for road selection of projects should include a poverty component. This can be a weighting by the size of population living in poverty within the zone of influence of a road, in combination with other conventional criteria such as population density and inter-linkages with other economic activities. The roads serving poverty areas could, for example, only be upgraded to a standard that is needed to connect them to the main road network.
In line with the above, a technical pre-feasibility analysis is required for projects to establish whether the identified areas are suitable for labour-intensive construction. The feasibility analysis would produce indicative information regarding the scope and scale of the project, together with the associated cost and time.
It is further recommended that communities and local community structures be fully involved at the planning stage and agreement should be reached based on, among others, the nature of the project, level of service, method of service delivery, availability of labour, selection of trainees and workers, wage rates and conditions of employment. Legitimate local-level institutions should be involved in the planning and implementation of projects within the Gundo Lashu programme; this will enhance greater community participation and ensure that the programme is assimilated into community development plans.
Disclosure statement
No potential conflict of interest was reported by the author.
References
- ADF (African Development Fund), 2007. Labour based public works appraisal report. [online]. http://www.afdb.org/fileadmin/uploads/afdb/Documents Accessed 15 March 2008.
- De Donder, P & Hendriks, J, 1998. The political economy of targeting. Public Choice 95, 177–200. doi: 10.1023/A:1005023531490
- Department of Public Works, 2007. Expanded public works programme. [online]. http://www.epwp.gov.za/ Accessed 3 July 2007.
- Department of Trasport, 2003. Rural transport strategy for South Africa. [online] www.transport.gov.za./RuralTransport20Strategy.pdf Accessed 12 November 2007.
- Elbers, C, Lanjouw, P, Mistiaen, J, Ozler, B & Simler, K, 2004. On the unequal inequality of poor communities. World Bank Research Observer 18(3), 401–21.
- Gelbach, JB & Pritchett, L, 1999. More for the poor is less for the poor: The politics of targeting. World Bank Development Research Group, World Bank, Washington, DC.
- Harmse, AC, 2004. Development regions in the South African space-economy: A multivariate approach. University of South Africa, Pretoria.
- Mashiri, M, Thevadasan, D & Zukulu, R, 2005. Community-Based labour –Intensive road construction: Findings of an impact study of the Amadiba Road. Proceedings of the 24th Southern African transport Conference (SATC), 11–13 July, Pretoria.
- McCord, A & Van Seventer, DE, 2004. The economy-wide impacts of the labour intensification of infrastructure expenditure in South Africa. Paper presented at the DPRU, TIPS & Cornell Conference on African Development and Poverty Reduction, the Macro-Micro Linkages, 13–15 October, Centre for Social Science Research, University of Cape Town, South Africa.
- McCutcheon, R, 1989. Labour-intensive road construction in Africa. Habitat International 13(4), 109–23. doi: 10.1016/0197-3975(89)90042-8
- McCutcheon, RT, 2008. Labour-intensive construction and maintenance in sub-Saharan Africa. A paper for discussion at the World Bank on the 7th April 2008. [online]. http://siteresources.worldbank.org Accessed 10 December 2008.
- Musekene, EN, 2013. The impact of a labour-intensive road construction programme in the Vhembe District, Limpopo Province. Development Southern Africa 30(3), 332–46. doi: 10.1080/0376835X.2013.817301
- Phillips, S, 2004. The expanded public works programme (EPWP). Presentation to the UNDP, HSRC & DBSA Conference on Overcoming Under-development in South Africa, 29 October, Pretoria.
- RAL (Road Agency Limpopo), 2009. Administrative datasets for Gundo Lashu programme. Limpopo province, Polokwane.
- Van Domelen, J, 2007. Reaching the poor and vulnerable: Targeting strategies for social funds and other community-driven programs. Human Development Network, World Bank.
- Van de Walle, D, 1999. “Public spending and the poor” Policy Research Working Paper No. 1476.
- Von Braun, J, Teklu, T & Webb, P, 1992. Labour-intensive public works for food security in Africa: Past experience and future potential. International Labour Review 131(1), 19–33.