- Email: [email protected]
The non-monetary quality of city life in South Africa
Habitat International 33 (2009) 319–326
Contents lists available at ScienceDirect
Habitat International journal homepage: www.elsevier.com/locate/habitatint
The non-monetary quality of city life in South Africa Wim Naude´ a, *, Stephanie Rossouw b, Waldo Krugell c a
World Institute for Development Economics Research, United Nations University, Katajanokanlaituri 6b, Helsinki, Finland Centre for Business Interdisciplinary Studies, Faculty of Business, Auckland University of Technology, New Zealand c School of Economics, North-West University, Potchefstroom, South Africa b
a b s t r a c t Keywords: Quality of life Urbanisation Cities South Africa JEL classification: R12 R23 O18 J67
In contrast to most research on the non-monetary quality of life, which relies on subjective indicators, we construct objective measures of the non-monetary quality of life using regression methods, for South Africa’s cities. We also analyse the extent to which the various cities have been able to turn improvements in per capita incomes (monetary quality of life) into non-monetary quality of life – as reflected for instance in a better environment, higher literacy and longer lives. When monetary quality of life measures are used for South Africa’s cities, the ranking in 2004 was led by Johannesburg, Tshwane, Ekurhuleni, Cape Town, Durban and Port Elizabeth. However when residuals from a regression of per capita income on (HDI) are used as a measure of non-monetary quality of life (i.e. the proportion of HDI not explained by variation in incomes), coastal cities tend to obtain generally higher rankings, with Cape Town ranked first, followed by Ekurhuleni, Durban, Port Elizabeth and then Johannesburg and Tshwane. Ó 2008 Elsevier Ltd. All rights reserved.
Introduction Unlike the rest of Africa, most people in South Africa live in cities. By 2005 an estimated 59.3% of its population was already living in urban areas, compared to the Sub-Saharan Africa average of 35% and the global average of 48.6% (United Nations, 2007). The UN, in its World Urbanization Prospects, predicts urbanisation to continue to increase in South Africa by around 1.35% per annum over the period 2005–2010. At least from a strict economic perspective this may be a welcome trend, as a strong case supported by empirical evidence can be made that urbanisation is good for economic productivity and income levels (see Naude´ & Krugell, 2003a, 2006; see also Rosenthal & Strange, 2004). It is therefore not surprising that also in South Africa there is a strong positive association between income level and urbanisation: average incomes per capita is already five times as high in cities than in rural areas. While urbanisation may be associated with higher average income levels, it is also often associated with a number of increasing social problems. In developing countries such as South Africa, these can put a particularly onerous burden on policy makers and development institutions. In South Africa in particular, the country’s growing urbanisation is still characterised by significant income inequality and seems to be accompanied by increases
* Corresponding author. E-mail addresses: [email protected] (W. Naude´), stephanie.rossouw@ aut.ac.nz (S. Rossouw), [email protected] (W. Krugell). 0197-3975/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.habitatint.2008.08.004
in crime, congestion and environmental degradation. This suggests that an appropriate framework from which to evaluate the success of urbanisation is the framework provided by the concept of quality of life (QoL). The question of what is perceived as quality of life has been the subject of much research since the 1970s. Key contributions have been made by Alkire (2002), Cummins (1996), Cummins, McCabe, Romeo, and Gullone (1994), Griffin (1991), Narayan, Chambers, Shah, and Petesch (2000), Nussbaum (1988, 2000), Qizilbash (1996) and Sen (1984, 1993, 1996). As noted by Sumner (2003), the evolution of the meaning and measurement of quality of life has broadened from an initial concern about income towards a multidimensional understanding of quality of life, wherein it recognised that material well-being, as measured by Gross Domestic Product (GDP) per capita, cannot alone explain the quality of life in a country. Thus, most current attempts to measure the quality of life through compiling indices include a wide range of measures – a detailed discussion falls outside the scope of this paper, but the interested reader is referred to some of the most well-known quality of life indices that have been proposed such as the Physical Quality of Life Index (Morris, 1979), the Human Development Index (HDI) (UNDP, 1990), the Quality of Life Index (Dasgupta & Weale, 1992), the Comprehensive Quality of Life Survey (Cummins et al., 1994), the Index of Economic Well-being (Osberg & Sharpe, 2000) and the Quality of Life Index constructed by Narayan et al. (2000). A shortcoming of these indices is that none has the ability to completely capture the multidimensional nature of quality of life nor do they capture the domain consisting of the environmental
320
W. Naude´ et al. / Habitat International 33 (2009) 319–326
quality, even though researchers such as Perrings (1998) have found a positive correlation between quality of place with quality of life, which can be based on the supposition that the physical location and surroundings play a deterministic role in quality of life. Given the above, the purpose of this paper is to make a threefold contribution towards applying the concept of the quality of life to city life in developing countries. First, we argue that the monetary benefits of increasing number of people living in cities should be seen against the possible non-monetary burdens of city life, thus arguing for a distinction between the monetary and non-monetary aspects of the quality of life to be emphasized. We argue that this is particularly needed when one considers the usefulness of a quality of life index to influence policy choices, in particular the ability of cities to turn monetary gains due to urbanisation into non-monetary gains. Second, we apply our arguments to South Africa, where we follow McGillivray’s (2005) proposed method to construct an objective index1 for the non-monetary quality of life for six of South Africa’s cities. This fills a gap in the literature on the quality of life in South Africa, which has alternately either focused on the monetary aspects of the quality of life, or attempted to measure the nonmonetary quality of life using subjective indicators.2 For instance, most research to date on South Africa’s cities have been on urban poverty, migration and squatting/urban sprawl (De Swardt, Puoane, Chopra, & du Toit, 2005; Ramutsindela, 2002; Saff, 2001), segregation and residential integration (e.g. Horn, 2002; Kotze & Donaldson, 1998; Prinsloo & Cloete, 2002; Saff, 1995; Turok, 2001), inner-city decline (Morris, 1999) urban governance and urban crime issues including the transformation to and from the ‘apartheid-city’ (e.g. Jenkins, 1997; Lemon, 1991; Williams, 2000) urban vulnerability (Nomdo & Coetzee, 2002), social justice (Visser, 2001) and urban empowerment (Lotter, 2002). Research on the nonmonetary aspects of city life has been relatively limited, and has so far been restricted to subjective measurements (Møller, 2001; Møller & Devey, 2003). And finally we make a small contribution towards attempting to incorporate the quality of place into a quality of life index, using various measures of geographical/environmental quality in South Africa’s cities. The paper is structured as follows. In Profile of South Africa’s Cities an overview of the selected cities’ socio-economic profiles (including economic- and non-economic indicators) will be provided. In The Non-Monetary Quality of Life of this paper the concept of non-monetary (non-economic) quality of life will be discussed and the methodology used in constructing a new index measuring the abovementioned quality of life will be introduced. Concluding Remarks provides a summary and some concluding remarks.
Profile of South Africa’s cities Location South Africa has 6 metropolitan municipalities namely the City of Cape Town, eThekwini (Durban), Ekurhuleni (East Rand), the City
1 The need for appropriate objective indicators of the non-monetary quality of life is due to (a) the often close correspondence between subjective and objective indicators (see e.g. Møller, 2004), and (b) the need for information about the ‘actual state of problems and the effects of attempts to solve these’ (Veenhoven, 2004: 21). 2 A detailed overview of research on the subjective quality of life falls outside the scope of this paper. However, it may be noted that the impetus to subjective quality of life research in South Africa was given by Møller and Schlemmer (1983) whose study that led to the development of the South African Quality of Life Trends Project. See also Møller (1999).
of Johannesburg, Nelson Mandela Metropole (Port Elizabeth) and the City of Tshwane (Greater Pretoria). The locations of these 6 cities are shown in Fig. 1. In Fig. 1, 4 agglomerations can be distinguished, one, which consists of 3 clustered cities (Johannesburg, Ekurhuleni, Tshwane) in the centre of the country and 3 smaller agglomerations on the coast (Cape Town, Nelson Mandela Metro and eThekwini Metro). The historical factors that determined the location and development of South Africa’s cities are discussed in greater detail from an economic perspective in Naude´ and Krugell (2003b). For present purposes it may be useful to provide a brief background on the geographical location of the country’s cities. The inland agglomeration of Johannesburg, which together with the City of Tshwane (the name of the metro area which includes Pretoria) can be seen as South Africa’s primate city (Naude´ & Krugell, 2003a). It was established following the discovery of significant gold and platinum deposits, the mining and distribution of which was energy and transport intensive, creating favourable infrastructure also for manufacturing development. Apart from the inland agglomeration around the Johannesburg– Tshwane complex the remainder of South Africa’s metropolitan areas is located at the coast as can be seen in Fig. 1. These coastal cities predate the establishment of the Johannesburg-agglomerations and developed as the result of the maritime nature of the country’s European colonisation. Thus cities such as Cape Town, Nelson Mandela Metro and eThekwini (which contains Durban) owe their existence to their locations in facilitating ocean transport between Europe and the East. Even today these cities retain an important influence on the South African economy, as South Africa ranks amongst the top 12 maritime nations in the world (Chasomeris, 2005). Socio-economic profiles In Table 1, some basic socio-economic indicators – mainly of economic well-being – in South Africa’s six metropolitan cities are reported. Before discussing the contents of Table 1, a word on the data utilised is in order. Data was obtained from 1996 to 2001 Census data of Statistics South Africa. The latter date is of the most recent census in South Africa. Little reliable and consistent data on a city-level is available for subsequent periods. Data on house prices was obtained from ABSA’s house price indicators (see www.absa.co. za) and data on the HDI, Gini-coefficient and environmental profiles was obtained from Global Insight’s Regional Economic Focus (REF). From Table 1, it can be seen that in 2001, there resided 14.5 million people (about 32% of the total) in South Africa’s 6 metropolitan cities. The city with the largest population is Durban (the eThekwini Metro) with just over 3 million people, followed by Cape Town with 2.9 million people. Although these coastal cities contain as individual cities the highest numbers of people, the map in Fig. 1 indicated a significant interior concentration of people consisting of 3 interlinking cities of Johannesburg, Ekurhuleni (East Rand) and City of Tshwane (Pretoria). Table 1 shows that 7.2 million people reside in this area, which if taken as a single socio-economic agglomeration, would constitute South Africa’s primate city (Naude´ & Krugell, 2003a). As far as density is concerned, Table 1 shows that although Johannesburg may not be the largest in terms of population, it has by far the greatest population density, with more than 2000 persons/km2. This is almost twice the density of the second most densely populated city, Durban. In terms of economic wealth, Table 1 shows that Cape Town and Johannesburg have the lowest poverty rates, with residents of Johannesburg displaying the largest disposable income. However, the Nelson Mandela Metro (Port Elizabeth) has the lowest overall
W. Naude´ et al. / Habitat International 33 (2009) 319–326
321
Fig. 1. Geographical location of South Africa’s six metropolitan cities.
income inequality (as measured by the Gini-coefficient) although it is the city with the highest poverty rate. It is also the city with the highest unemployment rate. Generally, the table suggests a close correlation between unemployment and poverty (the correlation coefficient between poverty and unemployment is 0.88), with the latter being the lowest in Cape Town and the highest in Nelson Mandela Metro. Losing one’s job or failing to find one in a South African city may therefore be a straight path to poverty. Table 2 contains some further indicators of economic well-being in South Africa’s metropolitan cities. It also contains the HDI, which
is a composite of economic indicators such as income as well as non-economic indicators such as life expectancy and literacy. In 2001, Johannesburg enjoyed the highest HDI, as well as the highest wage per worker (R 61,000 per annum or US $ 8777). Generally, wages in South Africa’s coastal cities appear to be lower than in the interior cities. In real terms, wages in all these cities contracted over the five years 1996–2001, with the largest contractions in Durban and Tshwane. The slowest contraction in wages was observed in Cape Town. Comparing the changes in real wage per worker with changes in unemployment, we find
Table 1 South Africa’s metropolitan cities in 2001: socio-economic profiles City
City of Cape Town eThekwini Metropolitan (Durban Unicity) Ekurhuleni Metropolitan (East Rand) City of Johannesburg Nelson Mandela Metropolitan (Port Elizabeth) City of Tshwane (Greater Pretoria) Total As % of South Africa a
South Africa’s cities in 2001: socio-economic status
Average household disposable incomea
Gini-coefficient
Population
Density
Poverty rate (%)
Unemployment rate (%)
2,954,774 3,077,928
582.91 1095.50
23.0 32.2
25.0 37.8
US $ 9107 US $ 8174
0.58 0.60
2,448,131
926.58
30.3
38.1
US $ 6792
0.58
2,672,006 1,078,477
2016.50 242.81
25.9 39.6
30.9 42.8
US $ 12 311 US $ 2514
0.60 0.57
2,294,632
410.51
30.6
29.7
US $ 8746
0.60
14,525,948 31.94
Converted from Rands to US $ using the average annual exchange rate of R 6.95 ¼ US $ 1.
US $ 47 645 51.30
W. Naude´ et al. / Habitat International 33 (2009) 319–326
322 Table 2 Socio-economic profile City (metropolitan government)
Average annual economic growth rate, 1996–2001 (%)
Wage per worker in 2001a
Annual average change in real wage per worker, 1996–2001 (%)
Human development index, 2001
Average house price, 2001a
City of Cape Town eThekwini Metropolitan (Durban Unicity) Ekurhuleni Metropolitan (East Rand) City of Johannesburg Nelson Mandela Metropolitan (Port Elizabeth) City of Tshwane (Greater Pretoria)
2.41 3.16
US $ 8058 US $ 7194
0.16 2.02
0.70 0.67
US $ 53 627 US $ 40 602
2.17
US $ 7050
0.81
0.67
US $ 39 505
4.60 4.58
US $ 8777 US $ 7482
0.88 0.61
0.72 0.66
US $ 48 348 US $ 38 497
5.22
US $ 7914
1.48
0.70
US $ 51 410
a
Converted from Rands to US $ using the average annual exchange rate of R 6.95 ¼ US $ 1.
a positive correlation of 0.68. Indeed, the city with the highest rate of increase in unemployment was Cape Town (a 28% increase in its unemployment rate) where increases in wages were highest, and the city with the lowest increase in unemployment was Tshwane (with an 11% increase) where increases in wages were the second lowest over the period. Table 2 also shows that the highest average annual economic growth rate over the period 1996–2001 was enjoyed by the City of Tshwane (Pretoria) of 5.22%, followed by Johannesburg and Nelson Mandela Metropolitan. House prices often reflect the underlying quality of life of a city, as well as a city’s economic importance or productivity (Rappaport & Sachs, 2003: 8). Table 2 shows that according to ABSA’s House Price information, in 2001 the highest average house price was reported in the City of Cape Town. This was followed by Pretoria and Johannesburg. Cape Town’s relatively lower average wage coupled with higher house prices may reflect the fact that it enjoys a higher quality of life due in part to its coastal location. The argument is that people are willing to accept lower wages and pay higher prices for housing in order to live in a coastal city (Rappaport & Sachs, 2003; Stover & Leven, 1992). The non-monetary quality of life Summary measures In the previous section one non-monetary indicator of quality of life, namely the HDI, was already mentioned. According to this measure, Johannesburg, Cape Town and Pretoria (Tshwane) enjoyed the highest standards of living. Another general indicator of the quality of life that reflects on both economic quality of life (such as due to higher productivity) as well as non-economic quality of life (such as scenery, climate, low crime, etc.) namely,
population density was also reported indirectly. According to Rappaport and Sachs (2003) population density within a country reflects the fact that, ‘people vote with their feet’. The previous section showed that population density is highest in Johannesburg, Durban and Ekurhuleni. Given that population density is highest in South African cities with higher wages per worker, higher growth and is generally located in the interior, leads to the working hypothesis that population density reflects high productivity (economic quality of life) rather than non-economic quality of life. Table 3 contains some implications for this hypothesis used. The overall crime index is significantly lower in the coastal city eThekwini (Durban) but it is significantly higher in Cape Town and Nelson Mandela Metro. When one considers the inland agglomeration, the city of Johannesburg has the highest overall crime index followed by Ekurhuleni and lastly the city of Tshwane. In Table 3, the population density has been included again to illustrate that there exists a slight positive correlation between population density and crime rates in South Africa’s cities. This reflects the fact that density, or agglomeration, not only benefits businesses through increasing returns to scale, but can also lead to criminal activity. The vehicle count, expressed as the number of vehicles per population, is clearly the highest in the Johannesburg – Ekurhuleni and Pretoria area, high in the Cape Town area, and much lower in the Durban and Nelson Mandela Metro area. Table 4 contains further measures of non-economic quality of life in South Africa’s metropolitan cities, in particular measures relating to the quality of the natural environment such as the percentage of a city’s surface area covered by forests, waterbodies and wetlands which may be indicative of natural beauty. In contrast, the percentage of degraded land would indicate poor environmental quality. The percentage of built-up land in a city could also provide a proxy indicator for the lack of open spaces and
Table 3 South Africa’s metropolitan cities in 2001: selected non-economic indicators South Africa’s cities in 2001: diverse non-economic indicators City (metropolitan government)
Overall crime indexa,b
City of Cape Town eThekwini Metropolitan (Durban Unicity) Ekurhuleni Metropolitan (East Rand) City of Johannesburg Nelson Mandela Metropolitan (Port Elizabeth) City of Tshwane (Greater Pretoria)
237.76 166.56 220.26 261.36 231.39 211.14
Total As % of South Africa
Vehicle count
Vehicles per person
Literacy rate (%)
831,516 449,358 686,565 880,686 190,216 666,339
0.28 0.15 0.28 0.33 0.18 0.29
86.2 82.6 83.1 87.1 83.7 82.9
3,704,680 51.70
0.26
Population
Density
2,954,774 3,077,928 2,448,131 2,672,006 1,078,477 2,294,632
582.91 1095.50 926.58 2016.50 242.81 410.51
14,525,948 31.94
Source of data: Global Insight Regional Economic Focus, November 2007. This index is the weighted average of crime/100,000 people. b This index was calculated using two separate approaches, namely the length-of-sentence approach and the cost-of-crime approach on defined violent crimes as well as property crimes. The final weights used in the Overall Crime Index were calculated as the average of the economic costs and the number of years in prison per 100,000 people. It should be noted that this index was chosen for two reasons: accuracy of reported crimes and ease of interpretation. For more information see www.globalinsight.co.za. a
W. Naude´ et al. / Habitat International 33 (2009) 319–326
323
Table 4 Environmental quality: selected indicators for 2004 City (metropolitan government)
Size (km2)
Forest, waterbodies & wetlands (%)
Degraded land (%)
Built-up land: residences (%)
Built-up land: commerce (%)
Mines (%)
City of Cape Town eThekwini Metropolitan (Durban Unicity) Ekurhuleni Metropolitan (East Rand) City of Johannesburg Nelson Mandela Metropolitan (Port Elizabeth) City of Tshwane (Greater Pretoria) Pretoria
5069 2810 2642 1325 4442 5590
4.2 2.3 3.5 2.3 1.4 38.2 22.8
6.23 6.97 0.00 0.00 0.10 17.41 0.00
11.28 20.60 23.45 55.81 4.59 16.58
1.29 2.30 3.61 6.01 0.85 1.07
0.15 0.07 4.49 4.10 0.12 0.22
Greater Pretoria’s significant % of land cover consisting of forest is especially found in Wonderboom (51%) and Ga-Rankuwa (43%) areas. Source of data: Global Insight Regional Economic Focus, November 2005.
access to nature; the percentage of residential buildings is also a proxy for access to housing. The choice of these variables was ultimately limited by data availability. It is common in indices of environmental quality to include CO2-emmissions (Zaim, 2005), which are unfortunately not available on the city-level in South Africa. From Table 4 it can be seen that the Greater Pretoria (Tshwane) area is endowed with the largest percentage of forests, waterbodies and wetlands than any city in South Africa – even if the central area of Pretoria is taken on its own, without the outlying areas such as Wonderboom and Ga-Rankuwa. Apart from Pretoria, it can be seen that Cape Town has the highest percentage of forests, waterbodies and wetlands – in addition to being a coastal city. It also displays some of the lowest percentage of build-up land, only lagging Nelson Mandela Metro in this regard. If the percentage of land area covered by mining operations can be judged to have a negative impact on environmental quality and thus quality of life, then our measures in Table 4 suggest that Ekurhuleni and Johannesburg are the most disadvantaged in this regard. The latter two cities are also the most densely built-up, with over 60 per cent of Johannesburg’s land area covered by residential and commercial buildings. Finally, Table 5 contains information on the climatic conditions in South Africa’s cities. We assume that quality of life is generally better is cities with higher rainfall and less variable annual temperature. Our measures indicate that climatic conditions are generally more favourable in South Africa’s coastal cities than in the noncoastal cities. For instance, rainfall in Cape Town and Durban tends to be higher than in most inland cities (except in this case, Ekurhuleni). Also, average annual temperatures are marginally higher along the coast, and the variations in annual temperature (between
Table 5 South Africa’s six metropolitan cities in 2001: climate City (metropolitan government)
South Africa’s Cities in 2001: climate Annual average rainfall (mm)
Average annual temperature ( C)
Variation in annual mean temperature
City of Cape Town eThekwini Metropolitan (Durban Unicity) Ekurhuleni Metropolitan (East Rand) City of Johannesburg Nelson Mandela Metropolitan (Port Elizabeth) City of Tshwane (Greater Pretoria)
683
17
3.03
Yes
939
21
2.77
Yes
703
16
3.99
No
655 502
16 18
3.81 3.07
No Yes
450
19
4.25
No
Coastal (yes/no)
highest and lowest average temperatures) are much less in coastal cities such as Durban, Cape Town and Nelson Mandela Metro. An own index for the non-monetary quality of life The United Nations’ Human Development Index (HDI), reported previously in Table 2 for South Africa’s six metropolitan cities, is one of the most widely used objective measures of non-monetary quality of life in use today. A major shortcoming of this measure is that it is highly correlated with per capita income. McGillivray (1991) finds a correlation coefficient of 0.89 between HDI and GNP per capita, and suggested that the HDI, as a true reflection of nonmonetary quality of life, may thus be ‘redundant’. In the present case, there is a similarly high positive correlation between per capita income and HDI across South Africa’s cities, of 0.84. A regression of the log of the HDI on the log of per capita income yielded the following estimates (see Table 6). To an extent the positive relationship between HDI and per capita income is due to the fact that per capita income is one component of the HDI – the other two being literacy and life expectancy. Thus, given that the HDI is not an exclusive indicator of the non-monetary quality of life (it contains per capita income) a non-monetary quality of life index was constructed. This index does not directly contain income or any other monetary aspects of quality of life. Due to data availability and ease of interpretation this approach follows the practice in Prescott-Allen (2001) by calculating a non-monetary quality of life index as the equally weighted average of measures of life expectancy, literacy and income equality. To ensure scale equivalence the components are expressed to range between 0 and 100.
Q1 ¼
Life expectancyþLiteracyþð1Gini coefficientÞ 3
(1)
In calculating Eq. (1), life expectancy is proxied by the percentage of the population in a city that is older than 75 years of age; the literacy rate is taken as the percentage of population older than 15 and functionally literate, and equality is measured by subtracting the Gini-coefficient from one. In Table 7 the index and its components are shown for South Africa’s six metropolitan cities for 2004.
Table 6 Regression results of (HDI) on per capita income across South Africa’s six metropolitan cities (data pooled for 1996, 2001 and 2004) Variable
Coefficient
Constant Per capita income Adj. R2 ¼ 0.81
1.87 (16.8)*** 0.146 (8.59)***
Dependent variable lnHDI. *** indicates statistical significance at the 5 % level.
W. Naude´ et al. / Habitat International 33 (2009) 319–326
324
Table 7 Index of the quality of life in South Africa’s six metropolitan cities, 2004 City
Cape Town eThekwini (Durban) Ekurhuleni (East Rand) Johannesburg Nelson Mandela Metro (Port Elizabeth) City of Tshwane (Pretoria)
Proportion of population older than 75 years of age (%)
Ginicoefficient
1.7 1.4 1.3 1.8 1.7
0.58 0.60 0.58 0.57 0.58
88 83 86 89 83
44 41 43 45 42
1.7
0.59
84
42
Literacy rate (%)
Quality of life index (non-Monetary)
Variable
Coefficient
Constant Per capita income Adj. R2 ¼ 0.12
1.32 (5.31)*** 0.05 (1.85)***
*** indicates statistical significance at the 5 % level.
Table 7 shows that in 2004 the non-monetary quality of life was highest in Johannesburg (45), Cape Town (44), Ekurhuleni (43) and lowest in Durban (41). The correlation between this composite index of the noneconomic quality of life and per capita income is still positive – with a correlation coefficient of 0.71. A regression of the log of this own measure on per capita income yields the results reported in Table 8. Regression analysis Given that both the HDI and the index of the non-monetary quality of life (Eq. (1)) are correlated with per capita income, regression analysis was used to obtain an even better measure of the non-monetary quality of life in South Africa’s metropolitan cities. The course followed by this paper has been initiated by McGillivray (2005) and is considered as the most progress to date in determining the non-economic quality of life. After briefly explaining the method used by McGillivray, this paper will follow suit to some degree in determining the non-economic quality of life for South Africa’s metropolitan cities. McGillivray (2005) extracted by means of principal component analysis, the maximum possible information from various standard non-economic quality of life achievement measures. He then empirically identified the variation in this extraction not accounted for variation in income per capita, which he named mi. This variable was then defined as being the residual yielded by cross-country regression of the extraction on the natural log of Purchasing Power Parity (PPP) GDP per capita. The variable mi can be interpreted inter alia as a measure of noneconomic human well-being/quality of life achievement, in the sense that it captures well-being achieved independently of income. For the purpose of this paper, this measure is obtained from the residuals as follows:
Qit ¼ a þ byit þ mit
Table 8 Regression results of a new index of non-monetary quality of life on per capita income across South Africa’s six metropolitan cities (data pooled for 1996, 2001 and 2004)
(2)
where Qit is the measure of quality of life in city i in period t (t ¼ 1996, 2001, 2004); HDI and the new index are used alternately; and yit is per capita income in city i in period t, with mit the residual term. The residual term can then be interpreted as the variance in Q (quality of life) that is not predicted by income per capita. We
therefore view this as a more appropriate or independent measure of the non-monetary quality of life in a city. McGillivray (2005: 340) also shows that this residual term can be interpreted as ‘a measure both of the success in converting economic well-being into noneconomic well-being and of the non-economic well-being component’. The results from estimating Eq. (2) with OLS using first HDI and second this paper’s new index (an own index for the non-monetary quality of life) are contained in Table 9. The residuals mit was in each case saved and used as an indicator. For the six metropolitan cities, the various non-monetary indicators of the quality of life (HDI, new indicator, HDI-residuals, new-residuals) as well as the contrast with the economic quality of life (per capita income) are shown in Table 9. In Table 9 the HDI, this paper’s newly constructed composite index of the non-monetary quality of life (consisting of the proportion of old age persons in the population, literacy and equality) as well as the residual estimates from Eq. (2) for HDI and this paper’s new index are shown. As indicated, the residuals can be interpreted as non-monetary quality of life indices that are independent of income, as well as indicators of the success with which the various cities are converting economic quality of life into nonmonetary quality of life. In order to make easier comparisons, the various cities in Table 9 have been ranked according to their per capita incomes. Although Johannesburg was ranked 1st in 2004 in terms of economic quality of life (using per capita income), it was only ranked 5th in terms of the residuals from the HDI, and 2nd in terms of the residuals from the new index constructed above. The City of Tshwane (Pretoria) is likewise ‘underperforming’ in terms of the non-monetary quality of life as measured by both the residuals from the HDI and this paper’s own index. Specifically, Tshwane is ranked the worst (6th) according to both measures. In contrast, the City of Cape Town is ranked 1st in South Africa on both estimates of the non-monetary quality of life, although in terms of per capita income it can only be ranked 4th in South Africa. Ekurhuleni (East Rand) and Durban’s performances seem to be on average: their per capita income ranking place them respectively in 3rd and 5th place, similar to their non-monetary quality of life rankings. Thus, it can be concluded that Pretoria and Johannesburg fare worst when it comes to non-monetary quality of life, but best when it comes to the economic quality of life. They also tend to be less successful when it comes to translating economic quality of life into non-monetary well-being. Cape Town, and to a lesser degree Nelson Mandela Metro (Port Elizabeth) fare better in terms of the non-monetary quality of life. They are also coastal cities. In the first
Table 9 Various non-monetary indicators of quality of life and rankings for South Africa’s metropolitan cities, 2004 City (metropolitan government)
HDI
Own index
(a) HDI residual
(b) New index residual
Per capita income
Ranking by (a)
Ranking by (b)
City of Johannesburg City of Tshwane (Greater Pretoria) Ekurhuleni Metropolitan (East Rand) City of Cape Town eThekwini Metropolitan (Durban Unicity) Nelson Mandela Metropolitan (Port Elizabeth)
0.73 0.70 0.68 0.70 0.67 0.65
0.4464 0.4229 0.4324 0.4369 0.4163 0.4225
0.009 0.013 0.001 0.023 0.003 0.005
0.011 0.007 0.010 0.015 0.005 0.005
US US US US US US
5 6 2 1 3 4
2 6 3 1 5 4
$ $ $ $ $ $
6178 5120 3810 3667 3480 2788
W. Naude´ et al. / Habitat International 33 (2009) 319–326
part of The Non-Monetary Quality of Life, it was shown that the overall crime index in South Africa’s cities was lowest in eThekwini (Durban) and highest in Johannesburg and that average climate and rainfall were also better in eThekwini (Durban) than elsewhere. As far as non-monetary quality of life is concerned, we deduce the metropolitan cities in South Africa that fare better are Cape Town and Nelson Mandela Metro. These are cities that fall in the middle and bottom categories of per capita income. The cities with a relatively lower non-monetary quality of life in South Africa are Tshwane (Pretoria), interestingly with one of the highest levels of per capita income, and Durban. Concluding remarks The paper confirms that it matters whether the quality of life is measured from a monetary or non-monetary point of view in South Africa’s metropolitan cities. When economic quality of life measures are used, specifically per capita income, the ranking in 2004 was led by Johannesburg, Tshwane, Ekurhuleni, Cape Town, eThekwini (Durban) and the Nelson Mandela Metro (Port Elizabeth). Clearly, the country’s largest agglomeration and its manufacturing base deliver the highest per capita income. When residuals from a regression of per capita income on HDI are used as a measure of non-monetary quality of life (i.e. the proportion of HDI not explained by variation in incomes), coastal cities tend to obtain generally higher rankings, with Cape Town ranked first, followed by Ekurhuleni, eThekwini (Durban), the Nelson Mandela Metro (Port Elizabeth) and then Johannesburg and Tshwane. A new composite index of non-monetary quality of life constructed in this paper resulted in a similar ranking, but with the major difference of finding that Johannesburg ranks in second place behind Cape Town. With urbanisation continuing apace in South Africa and the rest of Africa, the findings in this paper suggest that improvements in the non-monetary dimensions of well-being are important. Moreover, we suggest that this may be objectively measured, and policies may be informed to enable income gains in quality of life to be translated with more success into non-monetary outcomes. In the end, non-monetary quality of life is the ultimate objective in city development, with increases in incomes merely instruments to eventually achieve this objective. Acknowledgements Earlier versions of this paper were presented at the EcoMod International Conference on Regional and Urban Modelling at the Free University of Brussels, Belgium 2 June 2006 and at the 7th Conference of the International Society for Quality of Life Studies at Rhodes University, Grahamstown, South Africa, 17 July 2006. We are grateful to a number of participants for their useful comments and suggestions. We also wish to thank Thomas Gries for useful comments and suggestions on an earlier draft during an international workshop on regional development held at Paderborn University, Germany, on 6 June 2006. Last, but not least, the helpful comments and suggestions of an anonymous referee are gratefully acknowledged. The work benefitted from the gracious financial assistance of the Volkswagen Foundation, Germany. All errors and emissions are however, our own responsibility. References Alkire, S. (2002). Dimensions of human development. World Development, 30(2), 181–205. Chasomeris, M. (2005). Assessing South Africa’s shipping costs. Journal of Development Perspectives, 1(1), 125–141. Cummins, R. A. (1996). The domains of life satisfaction: an attempt to order chaos. Social Indicators Research, 38, 303–328.
325
Cummins, R. A., McCabe, M. P., Romeo, Y., & Gullone, E. (1994). The comprehensive quality of life scale: instrument development and psychometric evaluation on tertiary staff and students. Educational and Psychological Measurement, 54, 372–382. Dasgupta, P., & Weale, M. (1992). On Measuring the Quality of Life. World Development, 20(1), 119–131. De Swardt, C., Puoane, T., Chopra, M., & du Toit, A. (2005). Urban poverty in Cape Town. Environment and Urbanisation, 17(2), 101–112. Griffin, J. P. (1991). Against the taste model. In J. Elster, & J. Roemer (Eds.), Interpersonal comparisons of well-being (pp. 45–69). Cambridge: Cambridge University Press. Horn, A. C. (2002). New perspectives on urban segregation and desegregation in post-resolution South Africa. In I. Schnell, & W. Ostendorf (Eds.), Studies in segregation and desegregation (pp. 247–284). Aldershot: Ashgate. Jenkins, R. (1997). Rethinking ethnicity: Arguments and explorations. London: SAGE. Kotze, N. J., & Donaldson, S. E. (1998). Residential desegregation in two South African cities: a comparative study of Bloemfontein and Pietersburg. Urban Studies, 35, 467–477. Lemon, A. (1991). The apartheid city. In A. Lemon (Ed.), Homes apart: South Africa’s segregated city. Cape Town: David Philip. Lotter, J. C. (2002). Urban economic empowerment in the greater Pretoria area: the continuation of history? South African Journal of Economics, 70(2), 347–368. McGillivray, M. (1991). The human development index: yet another redundant composite development indicator? World Development, 19(10), 1461–1468. McGillivray, M. (2005). Measuring non-economic well-being achievement. The Review of Income and Wealth, 51(2), 337–364. Møller, V. (1999). South African quality of life trends in the late 1990s: major divides in perceptions. Society in Transition, 30(2), 93–106. Møller, V. (2001). Monitoring quality of life in cities: the Durban case. Development Southern Africa, 18(2), 217–238. Møller, V. (2004). Researching quality of life in a developing country: lessons from the South African case. Paper prepared for the Hanse Workshop on Researching Wellbeing in Developing Countries, 2–4 July 2004, Delmenhorst, Germany. Available from. http://www.welldev.org.uk/news/hanse-summaries/bremen.htm. Møller, V., & Devey, R. (2003). Trends in living conditions and satisfaction among poorer older South Africans: objective and subjective indicators of quality of life in the October Household Survey. Development Southern Africa, 20(4), 457–476. Møller, V., & Schlemmer, L. (1983). Quality of life in South Africa: towards an instrument for the assessment of quality of life and basic needs. Social Indicators Research, 12, 225–279. Morris, A. (1999). Tenant–landlord relations, the anti-apartheid struggle and physical decline in Hillbrow, an inner-city neighbourhood in Johannesburg. Urban Studies, 36, 509–526. Morris, D. (1979). Measuring the conditions of the world poor, the physical quality of life index. New York: Pergamon Press. Narayan, D., Chambers, R., Shah, M. K., & Petesch, P. (2000). Voices of the poor: Crying out for change. Washington, DC: World Bank. Naude´, W. A., & Krugell, W. F. (2003a). An enquiry into cities and their role in subnational economic growth in South Africa. Journal of African Economies, 12(4), 476–499. Naude´, W. A., & Krugell, W. F. (2003b). Are South African cities too small? Cities: The International Journal of Urban Policy and Planning, 20(3), 175–180. Naude´, W. A., & Krugell, W. F. (2006). Economic geography and growth in Africa: the case of sub-national convergence and divergence in South-Africa. Papers in Regional Science, 85(3), 443–457. Nomdo, C., & Coetzee, E. (2002). Urban vulnerability: Perspectives from Southern Africa. Sterling, VA: Stylus. Nussbaum, M. C. (1988). Nature, function and capability: aristotle on political distribution. Oxford Studies in Ancient Philosophy, 6, 145–184, Supplementary volume. Nussbaum, M. C. (2000). Women and human development: The capabilities approach. Cambridge: Cambridge University Press. Osberg, L., & Sharpe, A. International comparisons of trends in economic well-being. Paper presented at the annual meeting of the American economic association, January 2000, New York. Perrings, C. (1998). Income, consumption and human development: environmental linkages. Human development report. Background papers (pp. 151–212). Prescott-Allen, R. (2001). The wellbeing of Nations. Island Press/International Development Research Centre. Prinsloo, D. A., & Cloete, C. E. (2002). Post-apartheid residential mobility patterns in two South African cities. Property Management, 20(4), 264–277. Qizilbash, M. (1996). Capabilities, well-being and human development: a survey. Journal of Development Studies, 33(2), 143–162. Ramutsindela, M. (2002). Second time around: squatter removals in a democratic South Africa. GeoJournal, 57, 49–56. Rappaport, & Sachs, J. D. (2003). The United States as a coastal nation. Journal of Economic Growth, 8, 5–46. Rosenthal, S. S., & Strange, W. C. (2004). Evidence on the nature and sources of agglomeration economies. In J. V. Henderson, & J. F. Thisse (Eds.), Handbook of regional and urban economics, Vol. 4 (pp. 2119–2171). Amsterdam: Elsevier. Saff, G. R. (1995). Residential segregation in South Africa: what can be learned from the United States experience? Urban Affairs Review, 30, 782–808. Saff, G. R. (2001). Exclusionary discourse towards squatters in suburban Cape Town. Ecumene, 8, 87–107. Sen, A. K. (1984). Rights and capabilities. In A. K. Sen (Ed.), Resources, values and development. Oxford: Blackwell Press.
326
W. Naude´ et al. / Habitat International 33 (2009) 319–326
Sen, A. K. (1993). Capability and well-being. In M. C. Nussbaum, & A. K. Sen (Eds.), The quality of life. Oxford: Clarendon Press. Sen, A. K. (1996). On the foundations of welfare economics: utility, capability and practical reason. In F. Farina, F. Hahn, & S. Vannucci (Eds.), Ethics, rationality and economic behaviour. Oxford: Clarendon Press. Stover, M. E., & Leven, C. L. (1992). Methodological issues in the determination of the quality of life in urban areas. Urban Studies, 29(5), 737–754. Sumner, A. Economic and non-economic well-being: a review of progress on the meaning and measurement of poverty. Paper prepared for WIDER international conference on inequality, poverty and human well-being, May 30–31, 2003. Helsinki, Finland. Turok, I. (2001). Persistent polarization post-apartheid? Progress towards urban integration in Cape Town. Urban Studies, 38, 2349–2377.
UnitedNations. (2007). World urbanization prospects. http://esa.un.org/unup/. Accessed 01.08.08. United Nations Development Programme (UNDP). (1990). Human development report. New York: Oxford University Press. Veenhoven, R. (2004). Subjective measures of well-being. WIDER Discussion Paper no. 2004/07. Helsinki: United Nations University. Visser, V. (2001). Social justice, integrated development planning and post-apartheid South Africa. Urban Studies, 38(10), 1673–1699. Williams, J. (2000). South Africa – urban transformation. Cities, 17(3), 167– 183. Zaim, O. (2005). A framework for incorporating environmental indicators to the measurement of human well-being. WIDER Research Paper no. 2005/05. Helsinki: United Nations University.
Contents lists available at ScienceDirect
Habitat International journal homepage: www.elsevier.com/locate/habitatint
The non-monetary quality of city life in South Africa Wim Naude´ a, *, Stephanie Rossouw b, Waldo Krugell c a
World Institute for Development Economics Research, United Nations University, Katajanokanlaituri 6b, Helsinki, Finland Centre for Business Interdisciplinary Studies, Faculty of Business, Auckland University of Technology, New Zealand c School of Economics, North-West University, Potchefstroom, South Africa b
a b s t r a c t Keywords: Quality of life Urbanisation Cities South Africa JEL classification: R12 R23 O18 J67
In contrast to most research on the non-monetary quality of life, which relies on subjective indicators, we construct objective measures of the non-monetary quality of life using regression methods, for South Africa’s cities. We also analyse the extent to which the various cities have been able to turn improvements in per capita incomes (monetary quality of life) into non-monetary quality of life – as reflected for instance in a better environment, higher literacy and longer lives. When monetary quality of life measures are used for South Africa’s cities, the ranking in 2004 was led by Johannesburg, Tshwane, Ekurhuleni, Cape Town, Durban and Port Elizabeth. However when residuals from a regression of per capita income on (HDI) are used as a measure of non-monetary quality of life (i.e. the proportion of HDI not explained by variation in incomes), coastal cities tend to obtain generally higher rankings, with Cape Town ranked first, followed by Ekurhuleni, Durban, Port Elizabeth and then Johannesburg and Tshwane. Ó 2008 Elsevier Ltd. All rights reserved.
Introduction Unlike the rest of Africa, most people in South Africa live in cities. By 2005 an estimated 59.3% of its population was already living in urban areas, compared to the Sub-Saharan Africa average of 35% and the global average of 48.6% (United Nations, 2007). The UN, in its World Urbanization Prospects, predicts urbanisation to continue to increase in South Africa by around 1.35% per annum over the period 2005–2010. At least from a strict economic perspective this may be a welcome trend, as a strong case supported by empirical evidence can be made that urbanisation is good for economic productivity and income levels (see Naude´ & Krugell, 2003a, 2006; see also Rosenthal & Strange, 2004). It is therefore not surprising that also in South Africa there is a strong positive association between income level and urbanisation: average incomes per capita is already five times as high in cities than in rural areas. While urbanisation may be associated with higher average income levels, it is also often associated with a number of increasing social problems. In developing countries such as South Africa, these can put a particularly onerous burden on policy makers and development institutions. In South Africa in particular, the country’s growing urbanisation is still characterised by significant income inequality and seems to be accompanied by increases
* Corresponding author. E-mail addresses: [email protected] (W. Naude´), stephanie.rossouw@ aut.ac.nz (S. Rossouw), [email protected] (W. Krugell). 0197-3975/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.habitatint.2008.08.004
in crime, congestion and environmental degradation. This suggests that an appropriate framework from which to evaluate the success of urbanisation is the framework provided by the concept of quality of life (QoL). The question of what is perceived as quality of life has been the subject of much research since the 1970s. Key contributions have been made by Alkire (2002), Cummins (1996), Cummins, McCabe, Romeo, and Gullone (1994), Griffin (1991), Narayan, Chambers, Shah, and Petesch (2000), Nussbaum (1988, 2000), Qizilbash (1996) and Sen (1984, 1993, 1996). As noted by Sumner (2003), the evolution of the meaning and measurement of quality of life has broadened from an initial concern about income towards a multidimensional understanding of quality of life, wherein it recognised that material well-being, as measured by Gross Domestic Product (GDP) per capita, cannot alone explain the quality of life in a country. Thus, most current attempts to measure the quality of life through compiling indices include a wide range of measures – a detailed discussion falls outside the scope of this paper, but the interested reader is referred to some of the most well-known quality of life indices that have been proposed such as the Physical Quality of Life Index (Morris, 1979), the Human Development Index (HDI) (UNDP, 1990), the Quality of Life Index (Dasgupta & Weale, 1992), the Comprehensive Quality of Life Survey (Cummins et al., 1994), the Index of Economic Well-being (Osberg & Sharpe, 2000) and the Quality of Life Index constructed by Narayan et al. (2000). A shortcoming of these indices is that none has the ability to completely capture the multidimensional nature of quality of life nor do they capture the domain consisting of the environmental
320
W. Naude´ et al. / Habitat International 33 (2009) 319–326
quality, even though researchers such as Perrings (1998) have found a positive correlation between quality of place with quality of life, which can be based on the supposition that the physical location and surroundings play a deterministic role in quality of life. Given the above, the purpose of this paper is to make a threefold contribution towards applying the concept of the quality of life to city life in developing countries. First, we argue that the monetary benefits of increasing number of people living in cities should be seen against the possible non-monetary burdens of city life, thus arguing for a distinction between the monetary and non-monetary aspects of the quality of life to be emphasized. We argue that this is particularly needed when one considers the usefulness of a quality of life index to influence policy choices, in particular the ability of cities to turn monetary gains due to urbanisation into non-monetary gains. Second, we apply our arguments to South Africa, where we follow McGillivray’s (2005) proposed method to construct an objective index1 for the non-monetary quality of life for six of South Africa’s cities. This fills a gap in the literature on the quality of life in South Africa, which has alternately either focused on the monetary aspects of the quality of life, or attempted to measure the nonmonetary quality of life using subjective indicators.2 For instance, most research to date on South Africa’s cities have been on urban poverty, migration and squatting/urban sprawl (De Swardt, Puoane, Chopra, & du Toit, 2005; Ramutsindela, 2002; Saff, 2001), segregation and residential integration (e.g. Horn, 2002; Kotze & Donaldson, 1998; Prinsloo & Cloete, 2002; Saff, 1995; Turok, 2001), inner-city decline (Morris, 1999) urban governance and urban crime issues including the transformation to and from the ‘apartheid-city’ (e.g. Jenkins, 1997; Lemon, 1991; Williams, 2000) urban vulnerability (Nomdo & Coetzee, 2002), social justice (Visser, 2001) and urban empowerment (Lotter, 2002). Research on the nonmonetary aspects of city life has been relatively limited, and has so far been restricted to subjective measurements (Møller, 2001; Møller & Devey, 2003). And finally we make a small contribution towards attempting to incorporate the quality of place into a quality of life index, using various measures of geographical/environmental quality in South Africa’s cities. The paper is structured as follows. In Profile of South Africa’s Cities an overview of the selected cities’ socio-economic profiles (including economic- and non-economic indicators) will be provided. In The Non-Monetary Quality of Life of this paper the concept of non-monetary (non-economic) quality of life will be discussed and the methodology used in constructing a new index measuring the abovementioned quality of life will be introduced. Concluding Remarks provides a summary and some concluding remarks.
Profile of South Africa’s cities Location South Africa has 6 metropolitan municipalities namely the City of Cape Town, eThekwini (Durban), Ekurhuleni (East Rand), the City
1 The need for appropriate objective indicators of the non-monetary quality of life is due to (a) the often close correspondence between subjective and objective indicators (see e.g. Møller, 2004), and (b) the need for information about the ‘actual state of problems and the effects of attempts to solve these’ (Veenhoven, 2004: 21). 2 A detailed overview of research on the subjective quality of life falls outside the scope of this paper. However, it may be noted that the impetus to subjective quality of life research in South Africa was given by Møller and Schlemmer (1983) whose study that led to the development of the South African Quality of Life Trends Project. See also Møller (1999).
of Johannesburg, Nelson Mandela Metropole (Port Elizabeth) and the City of Tshwane (Greater Pretoria). The locations of these 6 cities are shown in Fig. 1. In Fig. 1, 4 agglomerations can be distinguished, one, which consists of 3 clustered cities (Johannesburg, Ekurhuleni, Tshwane) in the centre of the country and 3 smaller agglomerations on the coast (Cape Town, Nelson Mandela Metro and eThekwini Metro). The historical factors that determined the location and development of South Africa’s cities are discussed in greater detail from an economic perspective in Naude´ and Krugell (2003b). For present purposes it may be useful to provide a brief background on the geographical location of the country’s cities. The inland agglomeration of Johannesburg, which together with the City of Tshwane (the name of the metro area which includes Pretoria) can be seen as South Africa’s primate city (Naude´ & Krugell, 2003a). It was established following the discovery of significant gold and platinum deposits, the mining and distribution of which was energy and transport intensive, creating favourable infrastructure also for manufacturing development. Apart from the inland agglomeration around the Johannesburg– Tshwane complex the remainder of South Africa’s metropolitan areas is located at the coast as can be seen in Fig. 1. These coastal cities predate the establishment of the Johannesburg-agglomerations and developed as the result of the maritime nature of the country’s European colonisation. Thus cities such as Cape Town, Nelson Mandela Metro and eThekwini (which contains Durban) owe their existence to their locations in facilitating ocean transport between Europe and the East. Even today these cities retain an important influence on the South African economy, as South Africa ranks amongst the top 12 maritime nations in the world (Chasomeris, 2005). Socio-economic profiles In Table 1, some basic socio-economic indicators – mainly of economic well-being – in South Africa’s six metropolitan cities are reported. Before discussing the contents of Table 1, a word on the data utilised is in order. Data was obtained from 1996 to 2001 Census data of Statistics South Africa. The latter date is of the most recent census in South Africa. Little reliable and consistent data on a city-level is available for subsequent periods. Data on house prices was obtained from ABSA’s house price indicators (see www.absa.co. za) and data on the HDI, Gini-coefficient and environmental profiles was obtained from Global Insight’s Regional Economic Focus (REF). From Table 1, it can be seen that in 2001, there resided 14.5 million people (about 32% of the total) in South Africa’s 6 metropolitan cities. The city with the largest population is Durban (the eThekwini Metro) with just over 3 million people, followed by Cape Town with 2.9 million people. Although these coastal cities contain as individual cities the highest numbers of people, the map in Fig. 1 indicated a significant interior concentration of people consisting of 3 interlinking cities of Johannesburg, Ekurhuleni (East Rand) and City of Tshwane (Pretoria). Table 1 shows that 7.2 million people reside in this area, which if taken as a single socio-economic agglomeration, would constitute South Africa’s primate city (Naude´ & Krugell, 2003a). As far as density is concerned, Table 1 shows that although Johannesburg may not be the largest in terms of population, it has by far the greatest population density, with more than 2000 persons/km2. This is almost twice the density of the second most densely populated city, Durban. In terms of economic wealth, Table 1 shows that Cape Town and Johannesburg have the lowest poverty rates, with residents of Johannesburg displaying the largest disposable income. However, the Nelson Mandela Metro (Port Elizabeth) has the lowest overall
W. Naude´ et al. / Habitat International 33 (2009) 319–326
321
Fig. 1. Geographical location of South Africa’s six metropolitan cities.
income inequality (as measured by the Gini-coefficient) although it is the city with the highest poverty rate. It is also the city with the highest unemployment rate. Generally, the table suggests a close correlation between unemployment and poverty (the correlation coefficient between poverty and unemployment is 0.88), with the latter being the lowest in Cape Town and the highest in Nelson Mandela Metro. Losing one’s job or failing to find one in a South African city may therefore be a straight path to poverty. Table 2 contains some further indicators of economic well-being in South Africa’s metropolitan cities. It also contains the HDI, which
is a composite of economic indicators such as income as well as non-economic indicators such as life expectancy and literacy. In 2001, Johannesburg enjoyed the highest HDI, as well as the highest wage per worker (R 61,000 per annum or US $ 8777). Generally, wages in South Africa’s coastal cities appear to be lower than in the interior cities. In real terms, wages in all these cities contracted over the five years 1996–2001, with the largest contractions in Durban and Tshwane. The slowest contraction in wages was observed in Cape Town. Comparing the changes in real wage per worker with changes in unemployment, we find
Table 1 South Africa’s metropolitan cities in 2001: socio-economic profiles City
City of Cape Town eThekwini Metropolitan (Durban Unicity) Ekurhuleni Metropolitan (East Rand) City of Johannesburg Nelson Mandela Metropolitan (Port Elizabeth) City of Tshwane (Greater Pretoria) Total As % of South Africa a
South Africa’s cities in 2001: socio-economic status
Average household disposable incomea
Gini-coefficient
Population
Density
Poverty rate (%)
Unemployment rate (%)
2,954,774 3,077,928
582.91 1095.50
23.0 32.2
25.0 37.8
US $ 9107 US $ 8174
0.58 0.60
2,448,131
926.58
30.3
38.1
US $ 6792
0.58
2,672,006 1,078,477
2016.50 242.81
25.9 39.6
30.9 42.8
US $ 12 311 US $ 2514
0.60 0.57
2,294,632
410.51
30.6
29.7
US $ 8746
0.60
14,525,948 31.94
Converted from Rands to US $ using the average annual exchange rate of R 6.95 ¼ US $ 1.
US $ 47 645 51.30
W. Naude´ et al. / Habitat International 33 (2009) 319–326
322 Table 2 Socio-economic profile City (metropolitan government)
Average annual economic growth rate, 1996–2001 (%)
Wage per worker in 2001a
Annual average change in real wage per worker, 1996–2001 (%)
Human development index, 2001
Average house price, 2001a
City of Cape Town eThekwini Metropolitan (Durban Unicity) Ekurhuleni Metropolitan (East Rand) City of Johannesburg Nelson Mandela Metropolitan (Port Elizabeth) City of Tshwane (Greater Pretoria)
2.41 3.16
US $ 8058 US $ 7194
0.16 2.02
0.70 0.67
US $ 53 627 US $ 40 602
2.17
US $ 7050
0.81
0.67
US $ 39 505
4.60 4.58
US $ 8777 US $ 7482
0.88 0.61
0.72 0.66
US $ 48 348 US $ 38 497
5.22
US $ 7914
1.48
0.70
US $ 51 410
a
Converted from Rands to US $ using the average annual exchange rate of R 6.95 ¼ US $ 1.
a positive correlation of 0.68. Indeed, the city with the highest rate of increase in unemployment was Cape Town (a 28% increase in its unemployment rate) where increases in wages were highest, and the city with the lowest increase in unemployment was Tshwane (with an 11% increase) where increases in wages were the second lowest over the period. Table 2 also shows that the highest average annual economic growth rate over the period 1996–2001 was enjoyed by the City of Tshwane (Pretoria) of 5.22%, followed by Johannesburg and Nelson Mandela Metropolitan. House prices often reflect the underlying quality of life of a city, as well as a city’s economic importance or productivity (Rappaport & Sachs, 2003: 8). Table 2 shows that according to ABSA’s House Price information, in 2001 the highest average house price was reported in the City of Cape Town. This was followed by Pretoria and Johannesburg. Cape Town’s relatively lower average wage coupled with higher house prices may reflect the fact that it enjoys a higher quality of life due in part to its coastal location. The argument is that people are willing to accept lower wages and pay higher prices for housing in order to live in a coastal city (Rappaport & Sachs, 2003; Stover & Leven, 1992). The non-monetary quality of life Summary measures In the previous section one non-monetary indicator of quality of life, namely the HDI, was already mentioned. According to this measure, Johannesburg, Cape Town and Pretoria (Tshwane) enjoyed the highest standards of living. Another general indicator of the quality of life that reflects on both economic quality of life (such as due to higher productivity) as well as non-economic quality of life (such as scenery, climate, low crime, etc.) namely,
population density was also reported indirectly. According to Rappaport and Sachs (2003) population density within a country reflects the fact that, ‘people vote with their feet’. The previous section showed that population density is highest in Johannesburg, Durban and Ekurhuleni. Given that population density is highest in South African cities with higher wages per worker, higher growth and is generally located in the interior, leads to the working hypothesis that population density reflects high productivity (economic quality of life) rather than non-economic quality of life. Table 3 contains some implications for this hypothesis used. The overall crime index is significantly lower in the coastal city eThekwini (Durban) but it is significantly higher in Cape Town and Nelson Mandela Metro. When one considers the inland agglomeration, the city of Johannesburg has the highest overall crime index followed by Ekurhuleni and lastly the city of Tshwane. In Table 3, the population density has been included again to illustrate that there exists a slight positive correlation between population density and crime rates in South Africa’s cities. This reflects the fact that density, or agglomeration, not only benefits businesses through increasing returns to scale, but can also lead to criminal activity. The vehicle count, expressed as the number of vehicles per population, is clearly the highest in the Johannesburg – Ekurhuleni and Pretoria area, high in the Cape Town area, and much lower in the Durban and Nelson Mandela Metro area. Table 4 contains further measures of non-economic quality of life in South Africa’s metropolitan cities, in particular measures relating to the quality of the natural environment such as the percentage of a city’s surface area covered by forests, waterbodies and wetlands which may be indicative of natural beauty. In contrast, the percentage of degraded land would indicate poor environmental quality. The percentage of built-up land in a city could also provide a proxy indicator for the lack of open spaces and
Table 3 South Africa’s metropolitan cities in 2001: selected non-economic indicators South Africa’s cities in 2001: diverse non-economic indicators City (metropolitan government)
Overall crime indexa,b
City of Cape Town eThekwini Metropolitan (Durban Unicity) Ekurhuleni Metropolitan (East Rand) City of Johannesburg Nelson Mandela Metropolitan (Port Elizabeth) City of Tshwane (Greater Pretoria)
237.76 166.56 220.26 261.36 231.39 211.14
Total As % of South Africa
Vehicle count
Vehicles per person
Literacy rate (%)
831,516 449,358 686,565 880,686 190,216 666,339
0.28 0.15 0.28 0.33 0.18 0.29
86.2 82.6 83.1 87.1 83.7 82.9
3,704,680 51.70
0.26
Population
Density
2,954,774 3,077,928 2,448,131 2,672,006 1,078,477 2,294,632
582.91 1095.50 926.58 2016.50 242.81 410.51
14,525,948 31.94
Source of data: Global Insight Regional Economic Focus, November 2007. This index is the weighted average of crime/100,000 people. b This index was calculated using two separate approaches, namely the length-of-sentence approach and the cost-of-crime approach on defined violent crimes as well as property crimes. The final weights used in the Overall Crime Index were calculated as the average of the economic costs and the number of years in prison per 100,000 people. It should be noted that this index was chosen for two reasons: accuracy of reported crimes and ease of interpretation. For more information see www.globalinsight.co.za. a
W. Naude´ et al. / Habitat International 33 (2009) 319–326
323
Table 4 Environmental quality: selected indicators for 2004 City (metropolitan government)
Size (km2)
Forest, waterbodies & wetlands (%)
Degraded land (%)
Built-up land: residences (%)
Built-up land: commerce (%)
Mines (%)
City of Cape Town eThekwini Metropolitan (Durban Unicity) Ekurhuleni Metropolitan (East Rand) City of Johannesburg Nelson Mandela Metropolitan (Port Elizabeth) City of Tshwane (Greater Pretoria) Pretoria
5069 2810 2642 1325 4442 5590
4.2 2.3 3.5 2.3 1.4 38.2 22.8
6.23 6.97 0.00 0.00 0.10 17.41 0.00
11.28 20.60 23.45 55.81 4.59 16.58
1.29 2.30 3.61 6.01 0.85 1.07
0.15 0.07 4.49 4.10 0.12 0.22
Greater Pretoria’s significant % of land cover consisting of forest is especially found in Wonderboom (51%) and Ga-Rankuwa (43%) areas. Source of data: Global Insight Regional Economic Focus, November 2005.
access to nature; the percentage of residential buildings is also a proxy for access to housing. The choice of these variables was ultimately limited by data availability. It is common in indices of environmental quality to include CO2-emmissions (Zaim, 2005), which are unfortunately not available on the city-level in South Africa. From Table 4 it can be seen that the Greater Pretoria (Tshwane) area is endowed with the largest percentage of forests, waterbodies and wetlands than any city in South Africa – even if the central area of Pretoria is taken on its own, without the outlying areas such as Wonderboom and Ga-Rankuwa. Apart from Pretoria, it can be seen that Cape Town has the highest percentage of forests, waterbodies and wetlands – in addition to being a coastal city. It also displays some of the lowest percentage of build-up land, only lagging Nelson Mandela Metro in this regard. If the percentage of land area covered by mining operations can be judged to have a negative impact on environmental quality and thus quality of life, then our measures in Table 4 suggest that Ekurhuleni and Johannesburg are the most disadvantaged in this regard. The latter two cities are also the most densely built-up, with over 60 per cent of Johannesburg’s land area covered by residential and commercial buildings. Finally, Table 5 contains information on the climatic conditions in South Africa’s cities. We assume that quality of life is generally better is cities with higher rainfall and less variable annual temperature. Our measures indicate that climatic conditions are generally more favourable in South Africa’s coastal cities than in the noncoastal cities. For instance, rainfall in Cape Town and Durban tends to be higher than in most inland cities (except in this case, Ekurhuleni). Also, average annual temperatures are marginally higher along the coast, and the variations in annual temperature (between
Table 5 South Africa’s six metropolitan cities in 2001: climate City (metropolitan government)
South Africa’s Cities in 2001: climate Annual average rainfall (mm)
Average annual temperature ( C)
Variation in annual mean temperature
City of Cape Town eThekwini Metropolitan (Durban Unicity) Ekurhuleni Metropolitan (East Rand) City of Johannesburg Nelson Mandela Metropolitan (Port Elizabeth) City of Tshwane (Greater Pretoria)
683
17
3.03
Yes
939
21
2.77
Yes
703
16
3.99
No
655 502
16 18
3.81 3.07
No Yes
450
19
4.25
No
Coastal (yes/no)
highest and lowest average temperatures) are much less in coastal cities such as Durban, Cape Town and Nelson Mandela Metro. An own index for the non-monetary quality of life The United Nations’ Human Development Index (HDI), reported previously in Table 2 for South Africa’s six metropolitan cities, is one of the most widely used objective measures of non-monetary quality of life in use today. A major shortcoming of this measure is that it is highly correlated with per capita income. McGillivray (1991) finds a correlation coefficient of 0.89 between HDI and GNP per capita, and suggested that the HDI, as a true reflection of nonmonetary quality of life, may thus be ‘redundant’. In the present case, there is a similarly high positive correlation between per capita income and HDI across South Africa’s cities, of 0.84. A regression of the log of the HDI on the log of per capita income yielded the following estimates (see Table 6). To an extent the positive relationship between HDI and per capita income is due to the fact that per capita income is one component of the HDI – the other two being literacy and life expectancy. Thus, given that the HDI is not an exclusive indicator of the non-monetary quality of life (it contains per capita income) a non-monetary quality of life index was constructed. This index does not directly contain income or any other monetary aspects of quality of life. Due to data availability and ease of interpretation this approach follows the practice in Prescott-Allen (2001) by calculating a non-monetary quality of life index as the equally weighted average of measures of life expectancy, literacy and income equality. To ensure scale equivalence the components are expressed to range between 0 and 100.
Q1 ¼
Life expectancyþLiteracyþð1Gini coefficientÞ 3
(1)
In calculating Eq. (1), life expectancy is proxied by the percentage of the population in a city that is older than 75 years of age; the literacy rate is taken as the percentage of population older than 15 and functionally literate, and equality is measured by subtracting the Gini-coefficient from one. In Table 7 the index and its components are shown for South Africa’s six metropolitan cities for 2004.
Table 6 Regression results of (HDI) on per capita income across South Africa’s six metropolitan cities (data pooled for 1996, 2001 and 2004) Variable
Coefficient
Constant Per capita income Adj. R2 ¼ 0.81
1.87 (16.8)*** 0.146 (8.59)***
Dependent variable lnHDI. *** indicates statistical significance at the 5 % level.
W. Naude´ et al. / Habitat International 33 (2009) 319–326
324
Table 7 Index of the quality of life in South Africa’s six metropolitan cities, 2004 City
Cape Town eThekwini (Durban) Ekurhuleni (East Rand) Johannesburg Nelson Mandela Metro (Port Elizabeth) City of Tshwane (Pretoria)
Proportion of population older than 75 years of age (%)
Ginicoefficient
1.7 1.4 1.3 1.8 1.7
0.58 0.60 0.58 0.57 0.58
88 83 86 89 83
44 41 43 45 42
1.7
0.59
84
42
Literacy rate (%)
Quality of life index (non-Monetary)
Variable
Coefficient
Constant Per capita income Adj. R2 ¼ 0.12
1.32 (5.31)*** 0.05 (1.85)***
*** indicates statistical significance at the 5 % level.
Table 7 shows that in 2004 the non-monetary quality of life was highest in Johannesburg (45), Cape Town (44), Ekurhuleni (43) and lowest in Durban (41). The correlation between this composite index of the noneconomic quality of life and per capita income is still positive – with a correlation coefficient of 0.71. A regression of the log of this own measure on per capita income yields the results reported in Table 8. Regression analysis Given that both the HDI and the index of the non-monetary quality of life (Eq. (1)) are correlated with per capita income, regression analysis was used to obtain an even better measure of the non-monetary quality of life in South Africa’s metropolitan cities. The course followed by this paper has been initiated by McGillivray (2005) and is considered as the most progress to date in determining the non-economic quality of life. After briefly explaining the method used by McGillivray, this paper will follow suit to some degree in determining the non-economic quality of life for South Africa’s metropolitan cities. McGillivray (2005) extracted by means of principal component analysis, the maximum possible information from various standard non-economic quality of life achievement measures. He then empirically identified the variation in this extraction not accounted for variation in income per capita, which he named mi. This variable was then defined as being the residual yielded by cross-country regression of the extraction on the natural log of Purchasing Power Parity (PPP) GDP per capita. The variable mi can be interpreted inter alia as a measure of noneconomic human well-being/quality of life achievement, in the sense that it captures well-being achieved independently of income. For the purpose of this paper, this measure is obtained from the residuals as follows:
Qit ¼ a þ byit þ mit
Table 8 Regression results of a new index of non-monetary quality of life on per capita income across South Africa’s six metropolitan cities (data pooled for 1996, 2001 and 2004)
(2)
where Qit is the measure of quality of life in city i in period t (t ¼ 1996, 2001, 2004); HDI and the new index are used alternately; and yit is per capita income in city i in period t, with mit the residual term. The residual term can then be interpreted as the variance in Q (quality of life) that is not predicted by income per capita. We
therefore view this as a more appropriate or independent measure of the non-monetary quality of life in a city. McGillivray (2005: 340) also shows that this residual term can be interpreted as ‘a measure both of the success in converting economic well-being into noneconomic well-being and of the non-economic well-being component’. The results from estimating Eq. (2) with OLS using first HDI and second this paper’s new index (an own index for the non-monetary quality of life) are contained in Table 9. The residuals mit was in each case saved and used as an indicator. For the six metropolitan cities, the various non-monetary indicators of the quality of life (HDI, new indicator, HDI-residuals, new-residuals) as well as the contrast with the economic quality of life (per capita income) are shown in Table 9. In Table 9 the HDI, this paper’s newly constructed composite index of the non-monetary quality of life (consisting of the proportion of old age persons in the population, literacy and equality) as well as the residual estimates from Eq. (2) for HDI and this paper’s new index are shown. As indicated, the residuals can be interpreted as non-monetary quality of life indices that are independent of income, as well as indicators of the success with which the various cities are converting economic quality of life into nonmonetary quality of life. In order to make easier comparisons, the various cities in Table 9 have been ranked according to their per capita incomes. Although Johannesburg was ranked 1st in 2004 in terms of economic quality of life (using per capita income), it was only ranked 5th in terms of the residuals from the HDI, and 2nd in terms of the residuals from the new index constructed above. The City of Tshwane (Pretoria) is likewise ‘underperforming’ in terms of the non-monetary quality of life as measured by both the residuals from the HDI and this paper’s own index. Specifically, Tshwane is ranked the worst (6th) according to both measures. In contrast, the City of Cape Town is ranked 1st in South Africa on both estimates of the non-monetary quality of life, although in terms of per capita income it can only be ranked 4th in South Africa. Ekurhuleni (East Rand) and Durban’s performances seem to be on average: their per capita income ranking place them respectively in 3rd and 5th place, similar to their non-monetary quality of life rankings. Thus, it can be concluded that Pretoria and Johannesburg fare worst when it comes to non-monetary quality of life, but best when it comes to the economic quality of life. They also tend to be less successful when it comes to translating economic quality of life into non-monetary well-being. Cape Town, and to a lesser degree Nelson Mandela Metro (Port Elizabeth) fare better in terms of the non-monetary quality of life. They are also coastal cities. In the first
Table 9 Various non-monetary indicators of quality of life and rankings for South Africa’s metropolitan cities, 2004 City (metropolitan government)
HDI
Own index
(a) HDI residual
(b) New index residual
Per capita income
Ranking by (a)
Ranking by (b)
City of Johannesburg City of Tshwane (Greater Pretoria) Ekurhuleni Metropolitan (East Rand) City of Cape Town eThekwini Metropolitan (Durban Unicity) Nelson Mandela Metropolitan (Port Elizabeth)
0.73 0.70 0.68 0.70 0.67 0.65
0.4464 0.4229 0.4324 0.4369 0.4163 0.4225
0.009 0.013 0.001 0.023 0.003 0.005
0.011 0.007 0.010 0.015 0.005 0.005
US US US US US US
5 6 2 1 3 4
2 6 3 1 5 4
$ $ $ $ $ $
6178 5120 3810 3667 3480 2788
W. Naude´ et al. / Habitat International 33 (2009) 319–326
part of The Non-Monetary Quality of Life, it was shown that the overall crime index in South Africa’s cities was lowest in eThekwini (Durban) and highest in Johannesburg and that average climate and rainfall were also better in eThekwini (Durban) than elsewhere. As far as non-monetary quality of life is concerned, we deduce the metropolitan cities in South Africa that fare better are Cape Town and Nelson Mandela Metro. These are cities that fall in the middle and bottom categories of per capita income. The cities with a relatively lower non-monetary quality of life in South Africa are Tshwane (Pretoria), interestingly with one of the highest levels of per capita income, and Durban. Concluding remarks The paper confirms that it matters whether the quality of life is measured from a monetary or non-monetary point of view in South Africa’s metropolitan cities. When economic quality of life measures are used, specifically per capita income, the ranking in 2004 was led by Johannesburg, Tshwane, Ekurhuleni, Cape Town, eThekwini (Durban) and the Nelson Mandela Metro (Port Elizabeth). Clearly, the country’s largest agglomeration and its manufacturing base deliver the highest per capita income. When residuals from a regression of per capita income on HDI are used as a measure of non-monetary quality of life (i.e. the proportion of HDI not explained by variation in incomes), coastal cities tend to obtain generally higher rankings, with Cape Town ranked first, followed by Ekurhuleni, eThekwini (Durban), the Nelson Mandela Metro (Port Elizabeth) and then Johannesburg and Tshwane. A new composite index of non-monetary quality of life constructed in this paper resulted in a similar ranking, but with the major difference of finding that Johannesburg ranks in second place behind Cape Town. With urbanisation continuing apace in South Africa and the rest of Africa, the findings in this paper suggest that improvements in the non-monetary dimensions of well-being are important. Moreover, we suggest that this may be objectively measured, and policies may be informed to enable income gains in quality of life to be translated with more success into non-monetary outcomes. In the end, non-monetary quality of life is the ultimate objective in city development, with increases in incomes merely instruments to eventually achieve this objective. Acknowledgements Earlier versions of this paper were presented at the EcoMod International Conference on Regional and Urban Modelling at the Free University of Brussels, Belgium 2 June 2006 and at the 7th Conference of the International Society for Quality of Life Studies at Rhodes University, Grahamstown, South Africa, 17 July 2006. We are grateful to a number of participants for their useful comments and suggestions. We also wish to thank Thomas Gries for useful comments and suggestions on an earlier draft during an international workshop on regional development held at Paderborn University, Germany, on 6 June 2006. Last, but not least, the helpful comments and suggestions of an anonymous referee are gratefully acknowledged. The work benefitted from the gracious financial assistance of the Volkswagen Foundation, Germany. All errors and emissions are however, our own responsibility. References Alkire, S. (2002). Dimensions of human development. World Development, 30(2), 181–205. Chasomeris, M. (2005). Assessing South Africa’s shipping costs. Journal of Development Perspectives, 1(1), 125–141. Cummins, R. A. (1996). The domains of life satisfaction: an attempt to order chaos. Social Indicators Research, 38, 303–328.
325
Cummins, R. A., McCabe, M. P., Romeo, Y., & Gullone, E. (1994). The comprehensive quality of life scale: instrument development and psychometric evaluation on tertiary staff and students. Educational and Psychological Measurement, 54, 372–382. Dasgupta, P., & Weale, M. (1992). On Measuring the Quality of Life. World Development, 20(1), 119–131. De Swardt, C., Puoane, T., Chopra, M., & du Toit, A. (2005). Urban poverty in Cape Town. Environment and Urbanisation, 17(2), 101–112. Griffin, J. P. (1991). Against the taste model. In J. Elster, & J. Roemer (Eds.), Interpersonal comparisons of well-being (pp. 45–69). Cambridge: Cambridge University Press. Horn, A. C. (2002). New perspectives on urban segregation and desegregation in post-resolution South Africa. In I. Schnell, & W. Ostendorf (Eds.), Studies in segregation and desegregation (pp. 247–284). Aldershot: Ashgate. Jenkins, R. (1997). Rethinking ethnicity: Arguments and explorations. London: SAGE. Kotze, N. J., & Donaldson, S. E. (1998). Residential desegregation in two South African cities: a comparative study of Bloemfontein and Pietersburg. Urban Studies, 35, 467–477. Lemon, A. (1991). The apartheid city. In A. Lemon (Ed.), Homes apart: South Africa’s segregated city. Cape Town: David Philip. Lotter, J. C. (2002). Urban economic empowerment in the greater Pretoria area: the continuation of history? South African Journal of Economics, 70(2), 347–368. McGillivray, M. (1991). The human development index: yet another redundant composite development indicator? World Development, 19(10), 1461–1468. McGillivray, M. (2005). Measuring non-economic well-being achievement. The Review of Income and Wealth, 51(2), 337–364. Møller, V. (1999). South African quality of life trends in the late 1990s: major divides in perceptions. Society in Transition, 30(2), 93–106. Møller, V. (2001). Monitoring quality of life in cities: the Durban case. Development Southern Africa, 18(2), 217–238. Møller, V. (2004). Researching quality of life in a developing country: lessons from the South African case. Paper prepared for the Hanse Workshop on Researching Wellbeing in Developing Countries, 2–4 July 2004, Delmenhorst, Germany. Available from. http://www.welldev.org.uk/news/hanse-summaries/bremen.htm. Møller, V., & Devey, R. (2003). Trends in living conditions and satisfaction among poorer older South Africans: objective and subjective indicators of quality of life in the October Household Survey. Development Southern Africa, 20(4), 457–476. Møller, V., & Schlemmer, L. (1983). Quality of life in South Africa: towards an instrument for the assessment of quality of life and basic needs. Social Indicators Research, 12, 225–279. Morris, A. (1999). Tenant–landlord relations, the anti-apartheid struggle and physical decline in Hillbrow, an inner-city neighbourhood in Johannesburg. Urban Studies, 36, 509–526. Morris, D. (1979). Measuring the conditions of the world poor, the physical quality of life index. New York: Pergamon Press. Narayan, D., Chambers, R., Shah, M. K., & Petesch, P. (2000). Voices of the poor: Crying out for change. Washington, DC: World Bank. Naude´, W. A., & Krugell, W. F. (2003a). An enquiry into cities and their role in subnational economic growth in South Africa. Journal of African Economies, 12(4), 476–499. Naude´, W. A., & Krugell, W. F. (2003b). Are South African cities too small? Cities: The International Journal of Urban Policy and Planning, 20(3), 175–180. Naude´, W. A., & Krugell, W. F. (2006). Economic geography and growth in Africa: the case of sub-national convergence and divergence in South-Africa. Papers in Regional Science, 85(3), 443–457. Nomdo, C., & Coetzee, E. (2002). Urban vulnerability: Perspectives from Southern Africa. Sterling, VA: Stylus. Nussbaum, M. C. (1988). Nature, function and capability: aristotle on political distribution. Oxford Studies in Ancient Philosophy, 6, 145–184, Supplementary volume. Nussbaum, M. C. (2000). Women and human development: The capabilities approach. Cambridge: Cambridge University Press. Osberg, L., & Sharpe, A. International comparisons of trends in economic well-being. Paper presented at the annual meeting of the American economic association, January 2000, New York. Perrings, C. (1998). Income, consumption and human development: environmental linkages. Human development report. Background papers (pp. 151–212). Prescott-Allen, R. (2001). The wellbeing of Nations. Island Press/International Development Research Centre. Prinsloo, D. A., & Cloete, C. E. (2002). Post-apartheid residential mobility patterns in two South African cities. Property Management, 20(4), 264–277. Qizilbash, M. (1996). Capabilities, well-being and human development: a survey. Journal of Development Studies, 33(2), 143–162. Ramutsindela, M. (2002). Second time around: squatter removals in a democratic South Africa. GeoJournal, 57, 49–56. Rappaport, & Sachs, J. D. (2003). The United States as a coastal nation. Journal of Economic Growth, 8, 5–46. Rosenthal, S. S., & Strange, W. C. (2004). Evidence on the nature and sources of agglomeration economies. In J. V. Henderson, & J. F. Thisse (Eds.), Handbook of regional and urban economics, Vol. 4 (pp. 2119–2171). Amsterdam: Elsevier. Saff, G. R. (1995). Residential segregation in South Africa: what can be learned from the United States experience? Urban Affairs Review, 30, 782–808. Saff, G. R. (2001). Exclusionary discourse towards squatters in suburban Cape Town. Ecumene, 8, 87–107. Sen, A. K. (1984). Rights and capabilities. In A. K. Sen (Ed.), Resources, values and development. Oxford: Blackwell Press.
326
W. Naude´ et al. / Habitat International 33 (2009) 319–326
Sen, A. K. (1993). Capability and well-being. In M. C. Nussbaum, & A. K. Sen (Eds.), The quality of life. Oxford: Clarendon Press. Sen, A. K. (1996). On the foundations of welfare economics: utility, capability and practical reason. In F. Farina, F. Hahn, & S. Vannucci (Eds.), Ethics, rationality and economic behaviour. Oxford: Clarendon Press. Stover, M. E., & Leven, C. L. (1992). Methodological issues in the determination of the quality of life in urban areas. Urban Studies, 29(5), 737–754. Sumner, A. Economic and non-economic well-being: a review of progress on the meaning and measurement of poverty. Paper prepared for WIDER international conference on inequality, poverty and human well-being, May 30–31, 2003. Helsinki, Finland. Turok, I. (2001). Persistent polarization post-apartheid? Progress towards urban integration in Cape Town. Urban Studies, 38, 2349–2377.
UnitedNations. (2007). World urbanization prospects. http://esa.un.org/unup/. Accessed 01.08.08. United Nations Development Programme (UNDP). (1990). Human development report. New York: Oxford University Press. Veenhoven, R. (2004). Subjective measures of well-being. WIDER Discussion Paper no. 2004/07. Helsinki: United Nations University. Visser, V. (2001). Social justice, integrated development planning and post-apartheid South Africa. Urban Studies, 38(10), 1673–1699. Williams, J. (2000). South Africa – urban transformation. Cities, 17(3), 167– 183. Zaim, O. (2005). A framework for incorporating environmental indicators to the measurement of human well-being. WIDER Research Paper no. 2005/05. Helsinki: United Nations University.