Tourism impact, distribution and development: The spatial structure of tourism in the Western Cape province of South Africa

The structural dimensions of a country's tourism sector, and in particular the spatial structure of tourism production and consumption, relate closely to the nature and extent of the impact that tourism can have. This article examines the spatial characteristics of tourism in the Western Cape province, one of South Africa's foremost international tourist regions, and where its government seeks to use tourism as an instrument of development and socio-economic transformation. To understand how this could be effected it is necessary to understand the spatial distributional effects of tourism, and the underlying reasons for it. To this end the article examines the spatial structure of the provincial accommodation sector as evidenced in patterns of accommodation supply and tourist usage (demand); and trends in the nature, direction and distribution of public and private-sector tourism investments. The central argument is that tourism is geographically focused, with tourist activities concentrated in a few locales and sub-regions. This follows the general demographic and economic contours of the province. Yet trends in capital investments tend to reinforce the spatial concentration of tourism. Attempts by the government to spread tourism's benefits have not been too successful due to institutional and capacity deficiencies. Greater emphasis should be placed on developing domestic tourism.

1. Introduction

This article examines the spatial structure of tourism in the Western Cape province of South Africa, one of the country's premier tourist regions. It evaluates provincial policy towards tourism development in the context of the impact and distribution of tourism, patterns in its supply and usage, investment trends, and changes in these over time.

Like the national government, the provincial authorities are placing a high policy premium on expanding the tourism sector and using tourism as a pathway to development, empowerment and social transformation. Yet international experience shows that the purported economic and developmental advantages of tourism which are often ardently claimed are not automatic, and not without dispute. Tourism has several negative externalities attached to it: it is subject to seasonal fluctuations and high elasticities of demand and can therefore be an unstable source of income and employment (De Kadt, 1979; Mathieson & Wall, 1982; Brohman, 1996; Sinclair, 1998). It can also lead to higher inflation and balance-of-payments difficulties in destination countries (Britton, 1982; Mathieson & Wall, 1982; Lea, 1988).

Further, research indicates that three structural factors influence the economic impact of tourism: the strength of backward and forward linkages among industries (Sinclair & Tsegaye, 1990; Britton, 1991); the ownership structures and patterns that typify the sector; and the spatial features of production and related consumption in a given tourism economy (Britton, 1982; Opperman, 1993; Pearce, 1995). If weak linkages prevail among economic sectors and industries directly involved in tourism, the flow-through effect of tourism's impact, particularly from core to ancillary industries such as construction and transport will be weaker. Similarly, if tourism producer and supplier companies are not owned in the destination but in a different country, foreign exchange leakages, rather than earnings, may characterise tourism in the destination country (Brohman, 1996; Sinclair, 1998).

However, it is particularly the third factor – the spatial structure of tourism in a given destination – that relates closely to the specific nature of the economic gains that accrue from tourism and also to the extent of its impact in that destination (Opperman, 1993). More specifically, the location of tourist sites, the existence of tourism infrastructure and patterns in tourism production and consumption as manifested in, for instance, tourist circuits or travel flows, connect critically to the eventual impact of tourism. The manner in which people travel through a destination determines the types of interface that can take place between tourists and residents, and the potential for revenue creation among the latter. Further, if tourism production and consumption occur in spatially concentrated locales, its impact is similarly concentrated (Pearce, 1995). Much can be learned about the benefit that tourism has in a destination, by examining the geography and diffusion of its impact in terms of patterns of production and consumption. Even so, the structural and spatial aspects of tourism are often overlooked in the devising of tourism development policy (Pearce, 1995).

Several studies have been conducted on the geography of tourism supply in the Western Cape and in South Africa more broadly (Cape Town City Council, 1993, 1994; Deloitte & Touche, 1993; Visser & Van Huyssteen, 1997; Wesgro & KPMG, 1999). These, however, do not provide an analysis that focuses on the dynamic interplay between tourism production and consumption, the role that the state or industry may play in demand creation, for instance through tourism investments or infrastructure developments, and how this relates to development. To date, with a few exceptions on South Africa's tourism space economy (Demhardt, 2000; Rogerson, 2002; Visser, 2003), limited research has been conducted on the spatial dimensions of tourism in South Africa and the factors that contribute to it, and its role in enabling or constricting the government in achieving its transformation and developmental goals.

This article focuses on the spatial dynamics and the distribution of tourism impact in the Western Cape. It investigates the current structure of tourism in terms of location (ie supply or production) and demand (or consumption) factors and the interplay between these factors; what changes have taken place since 1994 and what underlay the changes; and what the role has been of key actors such as the provincial government and the private sector. Two particular facets are investigated:

	the geography of the accommodation sector as evidenced in patterns of accommodation supply and tourist usage (demand), and changes in these over time
	trends in the nature, direction and distribution of public and private sector tourism investments

The analysis shows that tourism activity, and concomitantly its impact, is geographically concentrated, being focused in certain parts of the Cape metropolis and in the Winelands and Garden Route. This is partly a consequence of the historical development of tourism in the province, arising from the promotion of a number of traditional, core attractions (or ‘icons’) and sites, and is related partly to the physical, demographic and infrastructural features of the province. Yet capital investments and promotion behaviour by both the government and the private sector (particularly in their focus on the international market) tend to reinforce, rather than significantly reshape, the current skewed structure. The government has some role to play in directing new development. At the same time provincial policy aptly aims at redistribution, but is offset by institutional and capacity deficiencies in tourism governance.

The first part of the article provides a brief overview of the Western Cape tourism sector, highlighting its prime attractions and the key tourism demand factors. The second part details the methodology that was used in assessing tourism's spatial structure. The third presents data on patterns in accommodation supply and demand since 1994 and large-scale investment trends. A fourth section discusses some of the wider implications of the geographic features for tourism policy in the province.

2. Western Cape tourism patterns and profiles

The Western Cape has flourished as one of the foremost international tourist destinations in the country over the past number of years. The province has several physical and other qualities that are important for its tourist appeal. Many of these (such as Table Mountain, Cape Point, the Garden Route and Kirstenbosch Botanical Gardens) have long been marketed internationally and are well-established ‘brand names’ and important tourism icons, not only for the Western Cape, but also the rest of the country (DEATT, 1999). This, along with the popularity of the province in general, means that it is one of South Africa's most important international tourist attractions. In line with this, and the national government's focus on expanding the tourism sector, boosting economic development through tourism has become a prime policy focus in the province (DEAAT, 2001).

One estimate of the size of the province's tourism economy, using Tourism Satellite Account methodology,1 put it at 13 per cent of the Gross Regional Product in 1996 (Bloom, 1998). A further estimate is that tourism provides employment to 9 per cent of the provincial population (Wesgro, 2001).

The Western Cape is divided into eight tourist regions that coincide roughly with the province's administrative districts. They are: the City of Cape Town; Winelands; Overberg; Garden Route; Klein Karoo Kannaland region; Central Karoo; Breede River Valley; and the West Coast.

The province has a significant share in South Africa's international tourism market, in 2000 receiving close to 60 per cent of all international visitors (Satour, 2000). The province is further the main destination for overseas tourists, drawing an average of 55 per cent of all overseas tourists in 1998 (Satour, 1997, 1998). In addition the province receives the lion's share of visitors from non-neighbouring Africa, in 1997 drawing 37 per cent of this component of South Africa's international market (Wesgro, 1999). These market profiles stand in sharp contrast to tourism in some of the Western Cape's main competitor provinces where, as in Gauteng, for example, business tourism or tourist arrivals from neighbouring African countries predominate. When one considers that the travel motivation and spending patterns of overseas visitors and from non-neighbouring Africa tend to be markedly different from that of business tourists – the former stay longer and spend more and generally come for the purpose of holiday – these market segments are highly significant: it indicates that the Western Cape draws more leisure tourism than any other province, and may be expected to draw a larger income from such tourism.

Survey data on domestic travel patterns in the province indicate that the City of Cape Town is the main destination for domestic tourists from other provinces (WCTB, 1999), receiving 72 per cent of all non-resident South African visitors. After this follow the Garden Route, the West Coast and the Winelands. The Klein and Central Karoo regions, Overberg and Breede River Valley receive only 1 per cent apiece of domestic visitors from the rest of South Africa (WCTB, 1999). Gauteng is the largest province of origin for domestic visitors. Most fall within the categories of ‘visiting friends and relatives’ or ‘holiday’. Considerably fewer people report travelling to the Western Cape for business purposes.

For domestic tourists resident in the Western Cape, the pattern is different, with the Garden Route and West Coast ranking as the most frequently visited regions (WCTB, 1999). In contrast to the preferred destinations of tourists from other provinces, regions such as the City of Cape Town and the Winelands receive relatively fewer Western Cape-resident tourists.

The Western Cape government, through its White Paper on sustainable tourism development and promotion (DEAAT, 2001), aims to foster tourism growth in the province. Specifically, it aims to increase the total number of overseas and domestic visitors to above 8 million, and to raise total tourist spending to above R14 billion by 2010 (DEAAT, 2001). It also seeks to develop a sustainable and inclusive tourism sector. Importantly, tourism is treated as a vehicle through which the economic and social transformation of the province could be achieved. In this respect, the chosen path for tourism development is one that emphasises:

	social equity through the promotion of accountable tourism management practices, and the development of tourist products that give equal representation to the peoples of the province
	economic empowerment through the creation of opportunities for ‘previously neglected communities’ to participate in mainstream tourism production and consumption, and by encouraging the growth of small-scale enterprises
	an integrated approach that seeks to enhance coordination among tourism sectors, and collaboration among the state, private sector, tourism employees and communities
	economic, institutional, environmental and social sustainability, defined as the existence of a diversity of market opportunities and employment, a judicious balance between the protection, and use of the environment, the participation of host communities in tourism management, and their cooperation with public and private sector actors (DEAAT, 2001)

The provincial policy is hence distinctive, in that it seeks to couple market and product expansion, aspects that are usually the focus of tourism development policies, with explicit socio-economic aims such as poverty alleviation and empowerment. In this way one of the key objectives of the Western Cape government is to assist in the restructuring of the tourism sector to spread the gains from tourism. By emphasising local ownership of tourism production and strong connections between economic sectors, the development framework is furthermore aiming to root tourism into the local economy.

However, in order to understand how tourism could be of developmental benefit for the residents of the Western Cape and, concomitantly, for the province's tourism development policy to deliver on its aims, it is necessary to understand the spatial distributional effects of tourism, and the underlying reasons for it.

3. Methodological note: Analysing the geographical impact and distribution of tourism

Broadly, tourism impact can be defined in terms of the number of tourists who visit a specific area, the amount of money that they spend, and the employment and income that is generated in the area as a consequence. Impact can also refer to the investments or production activities that either evolve from tourist activities in a given area, or are undertaken to prompt tourist activity (Smith, 1989; Sinclair, 1998). Tourism's impact, in other words, can be assessed through its direct and indirect effects. For this, demand-side or supply-side data can be utilised.

Demand-side studies of impact make use of indicators of effective demand, defined as the number of people who partake in a tourist activity or visit a tourist location (Pearce, 1995; Hall & Page, 1999). Common measures of effective demand are tourist expenditure (for example the sale of bed-nights in the hotel sector, or car rental days), visitor arrivals and departures at ports, and the duration of stay in a destination (Uysal, 1998). Studies can also assess employment related to tourism or calculate tourism's trade effects (the effect of tourism on the balance of payments) (Sinclair, 1998). More comprehensive methodologies seek to combine these various methods in order to derive at a measure of the overall impact that tourism has on a given economy.2

In contrast, geographical impact focuses on patterns of location, concentration (or spread) and flow. Specific analyses include assessments of the spatial configuration of tourism in a given region. This entails examining the location and distribution of facilities, attractions, accommodation stock, and so on. Another involves examining tourist movement within destinations. A third focuses on the geography of tourist usage of supply.

The analysis of the spatial structure and variations in tourism constitutes an important first step in understanding tourism in given area. The location of specific types of tourist facilities or attractions may be indicative of specific kinds of tourist activity that pertain to that area, and moreover may indicate the existence of certain tourist markets (Smith, 1989). The volume of supply (facilities or attractions) or of consumer usage in an area, and its relation to the resident population can be taken as a measure of the tourism level or intensity of that area (Defert, 1966; Keogh, 1984; Potts & Uysal, 1992). Finally, the distribution of infrastructure in a destination may account for patterns of tourist movement through that destination, and may be related to patterns of tourism impact (Pearce, 1995).

As tourism accommodation (such as hotels or guesthouses) is the most discernible component of tourism, accommodation statistics are most widely used to assess spatial variations in tourism (Pearce, 1995). Further, accommodation statistics constitute a most ready and visible measure of tourism impact in a local economy (Potts & Uysal, 1992), as the accommodation sector accounts for the largest proportion of tourist expenditure once tourists reach their destination. It also constitutes a useful surrogate indicator of two important features of tourism – the type and the level of tourist activity in a region (Pearce, 1995).

A set of quantitative and qualitative methods was used to assess the spatial distribution of tourism activity and impact in the Western Cape. On a province-wide (macro) level, three types of quantitative investigations were carried out. These were analyses of the level and location of tourism incidence as reflected in accommodation supply, tourist usage of accommodation and tourism investments. The purpose of this was to investigate both the static (locational) and dynamic (tourism flows or usage) elements of impact in the province, and to highlight geographical variations. In addition interviews were conducted with a sample of accommodation operators – a total of 40 establishments – and people in government and industry with interests in tourism.3

In order to assess patterns in accommodation demand, hotel trading statistics drawn from Statistics South Africa's regular survey of the hotel sector were analysed. Specifically, statistics on hotel bed-nights sold, and hotel income, were extracted for the years 1994 – chosen as base year as it was the year of South Africa's first democratic elections and when international tourism first increased – to 2001.

Data on accommodation supply was obtained from the Western Cape Tourism Board's (WCTB) accommodation directory for 2000–2001. This directory provides a listing of all accommodation establishments in the province (hotels, guesthouses, bed and breakfast establishments, backpackers and camping and caravan sites), and the number of rooms or beds in each establishment. The WCTB listing is the single most comprehensive directory on accommodation in the province. However, as it is new, it does not provide historical information on how accommodation supply has changed. For this, the accommodation listing of the Automobile Association of South Africa (AA) was used. The AA directory has been compiled since 1985 and provides similar information; it also contains information on accommodation establishments in different regions and towns and their capacity. Once again 1994 was chosen as the base year.

Data on tourism investments were extracted from the database on Major Investments and Development Opportunities (MIDO) compiled by the Western Cape Trade and Investments Promotion Agency (Wesgro). MIDO collates information on commercial, leisure/tourism and industrial/retail investments above a value of R10 million that are undertaken in the Western Cape. Such data are obtained from two main sources – local authorities, with whom applications for new developments have to be lodged, and the mass media.

MIDO is the most comprehensive database on large-scale investments in the province. A shortcoming is that because it includes only investments in excess of R10 million and generally does not incorporate smaller-scale expansion or refurbishment of existing projects, the database does not give a complete picture of all tourism development in the province. Indeed, tourism has been characterised by a significant degree of small-scale development, particularly in the accommodation sector through the voluminous growth of guesthouses and bed and breakfasts over the past number of years. While such developments are important, it is the capital investments that significantly reshape the fundamental structure of a tourism economy, and upon which smaller developments generally are contingent. Capital investments are usually a barometer of the direction of change in the underlying spatial structure of tourism in a given destination.

4. The spatial structure of the Western Cape accommodation sector

4.1 Overall accommodation supply

In 2001 there were more than 54 000 tourist beds in the province. Significantly, close to three-quarters of these were in only two types of establishments – hotels and self-catering establishments. Guesthouses and bed and breakfast establishments respectively provided about one-tenth of tourist beds, while backpackers and camping and caravan establishments together provided only 6 per cent of total accommodation in the province. As far as the availability of tourist beds is concerned, therefore, hotels and self-catering establishments are by far the most important components of the provincial accommodation sector. Backpackers and camping and caravan sites do not constitute a significant element of the accommodation supply. Figure 1 shows the relative share of tourist beds held by different accommodation types in each of the province's regions.

Figure 1: Regional distribution of accommodation supply in the Western Cape, 2001

The location and patterns of distribution of the various accommodation types indicate certain kinds of tourist activity that pertain to the different regions of the province. First, tourist activity geared towards hotel accommodation is concentrated in the four coastal regions, and is comparatively less often found in the rural regions of the Western Cape. Second, the self-catering sector is widely spread throughout the province, but occurs particularly in the rural areas. It may indicate that there is relatively robust tourist demand for this type of accommodation, ahead of other accommodation types such as guesthouses or bed and breakfasts.

As can clearly be seen, the City of Cape Town has the largest share of the accommodation stock. The City may therefore be said to hold particular importance in the provincial accommodation sector. Within the City itself, however, even though there has been general growth across all of the six municipal regions, the Cape Town municipality, and specifically more affluent parts such as the City Bowl and Atlantic Seaboard make up two-thirds of the total accommodation stock.

4.2 The provincial and metropolitan hotel sector

Tables 1 and 2 illustrate how hotels are distributed in the province and City. The spread of hotels follows the general pattern of accommodation supply in the Western Cape. The hotel sector is very significant in Cape Town, making up 65 per cent of total bed supply in the city. Again, within the City, there is a concentration of hotel units in the Cape Town municipality.

Table 1: Hotel distribution in the Western Cape, 2001

Region	Number of hotel units	Per cent of total	Number of beds	Per cent of total	Average number of beds per hotel unit	Hotel beds as percentage of total number of beds in region
City of Cape Town	100	47,4	13 523	64	135	65
Garden Route	42	19,9	4 020	19	96	33
Overberg	21	10,0	1 067	5	51	17
West Coast	20	9,5	704	3,3	35	14
Breede River Valley	11	5,2	533	2,5	49	13
Winelands	7	3,3	424	2	61	15
Klein Karoo	7	3,3	574	2,7	82	22
Central Karoo	3	1,4	302	1,4	101	42
Total	211		21 147		100
Source: Western Cape Tourism Board (2001).

Table 2: Distribution of hotel beds in the Cape metropolis, 2001

Municipality	Number of hotel units	Number of hotel beds	Per cent of total	Average number of beds per hotel
Cape Town	67	10 812	80	161
Tygerberg	9	1 076	8	119
Helderberg	9	779	5,8	86
South Peninsula	9	360	2,6	40
Blaauwberg	4	280	2,0	70
Oostenberg	2	216	1,6	108
Total	100	13 523
Source: Western Cape Tourism Board (2001).

Between 1994 and 2001 significant changes have taken place in the city's hotel sector. There has been a great expansion of the city's hotel stock during these years. The greatest change has however taken place only after 1996, when within five years the number of hotel beds has nearly doubled. A large proportion of hotel expansion took place between 1996 and 1998/1999. For instance, 32 new hotels were developed in the City of Cape Town between May 1996 and May 1998 – 13 hotels in 1996, 15 in 1997 and four in 1998 (Wesgro, 1998).

Interviews with people active in the tourism industry indicated that very little hotel development took place in the province's rural areas. The growth in the metropolitan hotel sector was attributed to a number of factors. First, double-digit growth in international tourist arrivals in the aftermath of South Africa's first democratic elections was said to have led to a rise in demand for hotel accommodation. Second, the success of the Rugby World Cup in 1995 led to optimism over the future of the South African and Western Cape tourism industries in the mid-1990s. The bid process for the Olympic Games in 2004 which started in the early 1990s set in motion a series of state and privately-funded development projects in the greater metropolis, which continued even after the city failed to be awarded the games (Hiller, 2000). These factors had the effect that investment in the hotel sector once again became viable after the downturn of the sanctions-era of the 1980s.4 Within this context, many new hotel investments were made under a tax incentive scheme that had been introduced by the former government.5 According to knowledgeable informants, another reason why hotel investments expanded so rapidly during this period was the rise of sectional title developments, in which hotels are purpose-built for multiple ownership. In the middle of the 1990s, the rising popularity of this type of investment meant that the number of hotels in the city rapidly increased. During this time the provincial government also implemented new policies aimed at regulating the casino industry. While the establishment and licensing of three new casino complexes in the province (one each in the City of Cape Town, Overberg and West Coast) also entailed the development of hotels tied to such complexes, this did not have a large tourism impact according to most of those interviewed.6

4.3 Patterns in demand in the provincial hotel sector

In terms of tourist demand, a number of shifts had taken place in the hotel market of the Western Cape over the past decade. In the first instance, as can be seen in Figure 2, tourist usage of hotel accommodation has consistently grown. For the Western Cape demand for hotel accommodation has increased to the extent that on a national basis, the province has established a strong position, and takes one-quarter of the overall hotel market.

Figure 2: Trends in hotel bed-nights sold in the Western Cape, 1992–2001

Within the Western Cape itself, domestic tourism forms the mainstay of the province's hotel market, accounting for close to three-quarters of total provincial bed-nights sold, although by 1997 the province had secured a significant share of the national market for international tourists (approximately 42 per cent) (Figure 3).

Figure 3: The hotel market for international tourists in the Western Cape

Figure 4 and Table 3 illustrate the disparities between the hotel markets of the province's various sub-regions in terms of the volume of international tourism they draw, and tourist usage by star grading.

Figure 4: Share of hotel bed-nights sold to foreigners in the Western Cape, by region, 1995–1997

Table 3: Tourist usage, by star grading

Star grading	Ungraded	One- and two-star	Three-star	Four-star	Five-star	Three- to five-star	One- to four-star	Total
Per cent of bed nights sold, by star grading
City of Cape Town		13	41	28	18			100
Klein and Central Karoo	6	48	46					100
Garden Route	41	7	52					100
Winelands	28					72		100
West Coast	35						65	100
Breede River Valley	69						31	100
Overberg							100	100
Per cent of hotel income, by star grading
City of Cape Town		8	23	27	42			100
Klein and Central Karoo	25	29	46					100
Garden Route	19	13	68					100
Winelands	15					85		100
West Coast	37						63	100
Breede River Valley	43						57	100
Overberg							100	100
Source: Statistics South Africa (January 2001).

Table 3 presents data on the share of hotel bed-nights sold and hotel income generated in various star grading categories in the Western Cape's eight tourism regions during January 2001. It shows interesting contrasts in the relative importance of different types of hotels in different sub-regions. In Cape Town, for instance, it can be seen that tourism demand is highest in the three- and four-star sectors, with bed-nights sold in the five-star sector making up less than one-fifth of the total number of bed-nights sold. However, as far as hotel revenue is concerned, the five-star sector is the single most important earner in the city. It can also be seen that one- and two-star hotels do not play such a significant role in the city's hotel market. Similar features characterise the hotel markets of the Garden Route and Winelands. It is clear, however, that the ungraded hotel sector plays a very large role in the rest of the Western Cape province.

Overall, the City of Cape Town, the Winelands and the Garden Route are distinguished by the fact that they draw more international tourists, and more higher-yielding clients, than other parts of the Western Cape. Concomitantly, in these regions a major share of hotel impact is afforded through international tourism. In contrast, in the outlying rural areas of the Western Cape (the West Coast, Breede River Valley and Karoo regions) domestic tourism is the mainstay of the hotel sector.

Qualitative analyses of hotel demand indicate that increases in hotel supply in the province have not been met with sufficient levels of tourist usage. This has had attendant effects on occupancy levels. Interviews with hotel operators and managers in different areas of the Western Cape (a total of 20 hotels) show that the rise in hotel supply during the mid- to latter-1990s had substantially affected the hotel sector. Operators agreed that the sharp increase in international tourist arrivals had translated into rising demand for accommodation. Until 1996, ‘business was booming’. By 2001, however, the situation had reversed: the growth in international visitors had levelled off, while supply had continued to increase. At present, therefore, operators agreed that demand was inadequate to sustain the hotel industry. Most operators spoke of a decided ‘oversupply’ in the province's hotel sector. One hotel executive stated: ‘We've got occupancy sitting at 50–60 per cent at the moment, which is really an industry operating at half capacity, it's not really a healthy industry.’7

International literature and empirical studies on the hotel property market suggest a general lag between hotel investments and occupancies, often because hotel investments tend to be made independent of macro-economic fluctuations and based on the price structure within the hotel sector (for example, Wheaton & Rossoff, 1997).

Lower occupancy levels in the metropolitan hotel sector could be a manifestation of the ‘normal’ lag between investments and occupancy. It needs to be asked, however, whether other deeper-lying or structural factors account for lower occupancy rates. South Africa has ostensibly gained from a destination substitution effect in the aftermath of the series of terrorist attacks of 2001 and 2002 and the unrelated, substantial weakening of the Rand at the end of 2002. For example, in contrast to a global tourism decline of 1 per cent in 2001 (WTO, 2002), international arrivals into South Africa grew by 11 per cent during the first part of 2001 and 2002 (DEAT Media Release, 9 March 2003). This had translated into an increase in hotel occupancies for 2002, to a national average of 57 per cent for the period June to July. In Cape Town, average hotel occupancy for this period was 63 per cent.8 Notwithstanding, the country's sector is subject to marked volatility. By the end of 2003, for instance, international arrivals had once again waned due to broad economic downturns in South Africa's main source markets. This, coupled with the strengthening in the country's currency, has had a downward effect on hotel occupancies (Business Day, 29 October 2003). In the City of Cape Town occupancies for the period between October 2002 and 2003 declined by seven per cent (Business Day, 2 December 2003).

Such volatility is compounded by a further negative characteristic – the Western Cape's tourism market is particularly subject to seasonal shifts, and seasonal variation in demand greatly affects the city's hotels. Occupancy generally peaks in January and February, drops off during March and April to low levels in June, July and August, before rising again during the spring months of October and November.

Hotel operators argued that within the context of over-capacity in the metropolis, the negative effect of seasonal demand on the hotel industry became more intense. Furthermore, these factors set the context for the operations, practices and market traits of the broad accommodation sector. There is extensive divergence between the importance of different market segments. Overall the corporate and the group traveller segments, which are less subject to seasonal fluctuations in demand, carry the greatest significance in the province's accommodation business, while the individual leisure traveller segment carries the least. Given the effect of seasonality, however, variability in market composition is an important feature. Generally the use of accommodation by different market segments interchanges over different periods of the year. Accommodation operators alter marketing schedules and market targeting in accordance with seasonal changes in demand. As a consequence of high levels of supply and intense competition, moreover, developing niche markets and alternative sources of income is a vital aspect of operators' activities.

Analysis of the Western Cape accommodation sector brings to light a number of key aspects. There has been a broad increase in accommodation supply after 1994, partially brought about by increasing demand after this period. This has produced a differential effect on different parts of the province and different components of the accommodation sector. The supply and tourist usage patterns show that different tourist functions and activities prevail in various parts of the province. There is a clear functional division based on types of markets: in the rural hinterland the domestic (and leisure) tourist market is most important, while in the Cape metropolis international tourism and corporate meetings, incentives, conferences and exhibitions (MICE), both higher-yielding sectors, play a very important role. Tourism intensity, and the importance of tourism, is also unevenly spread throughout the province. Broadly, most tourism takes place in the City of Cape Town (and specifically in the City Bowl and Atlantic Seaboard area), the Garden Route and, to a lesser extent, the Winelands. Concomitantly, accommodation impact, as reflected by actual tourist usage and income generated, tends to be concentrated in these areas.

The next section investigates trends in Western Cape tourism investment in excess of R10 million, between 1998 and 2001. The objective of this section is to examine the direction of tourist development and of demand stimulation in the province since 1998. These trends are related with patterns of accommodation supply and usage highlighted previously, and conclusions are drawn on the implications for future tourism growth and distribution.

5. Tourism investment patterns in the Western Cape, 1998–2001

Figures 5 and 6 present all the major tourism-specific and tourism-related investment projects undertaken in the Western Cape and Cape Town since 1998. Table 4 below presents a further breakdown of these projects by their status and regional location.

Figure 5: Distribution of tourism investments in the Western Cape, 1998–2001

Figure 6: Distribution of tourism investments in Cape Town, 1998–2001

Table 4: Tourism investments in the Western Cape, 1998–2001

Region	Total number of projects	Under construction (number)	Approved (number)	Planned (number)	Total known value of projects (R billion)
Cape Town	32	17	5	10	7.8
Winelands	11	4	3	4	1.95
Garden Route	9	8	–	1	5
Overberg	9	5	–	4	1.95
West Coast	6	3	1	2	0.31
Breede River Valley	3	–	–	3	0.15
Klein Karoo	2	–	–	2	0.005
Central Karoo	0	0	0	0	0
Total	72	37	9	26	R17.2bn
Source: Wesgro, MIDO database.

The figures and table display a strong degree of investment concentration in the Western Cape. Investments are most dense in the metropolitan area. Outside the city the Garden Route, the Winelands and the Overberg are the prime investment regions, while a number of projects are taking place along the West Coast. In the remainder of the province very little large-scale tourist development is occurring.

In other words, there is a high correlation between the investment and accommodation supply and usage patterns in the province. This has two implications. On the one hand, major tourism investments follow general levels of tourist activity, with investments being most focused where tourist activity is most intense. On the other hand, there appears to be very little in the way of large-scale demand stimulation in the regions where tourist activity is less dynamic, such as the Breede River Valley and the Karoo regions.

In part this pattern of capital investments stems from the demographic and economic structure of the province. The Western Cape is highly urbanised, with two-thirds of the population residing in the City of Cape Town. The metropolis also accounts for three-quarters of the total Gross Regional Product. A further analysis of the kinds of investments that are made in the Western Cape provides important insights into the direction of tourist development. Table 5 presents a breakdown of the province's investment projects by investment type and location.

Table 5: Breakdown of provincial tourism investments, by type and location

Primary investment category	Total number of projects	Location
		Metropolitan	Non-metropolitan
		1. Accommodation
a) Hotel	10	Helderberg;	Overberg (2)
		Blaauwberg;	West Coast
		Tygerberg (2)
b) Hotel + conference	4	Cape Town (3)	Overberg
		Cape Town	Winelands (2)
c) Hotel and residential	2		West Coast
d) Tourist accommodation	2	Cape Town	Klein Karoo
		Robben Island
2. Golf
a) Golf courses	5	Tygerberg	Winelands (3)
			Garden Route
b) Golf complexes (golf + residential or golf + hotel)	12	Blaauwberg	Overberg (2)
			Winelands (4)
			Garden Route (5)
3. Combination recreation/leisure and residential	10	Cape Town (2)	Garden Route (2)
		Tygerberg (3)	Wineland
		Blaauwberg	Overberg
4. Combination recreation/leisure and commercial	3		Breede River Valley
			West Coast
			Garden Route
5. Infrastructure	10	Tygerberg (5)	Winelands
		Cape Town (3)	Klein Karoo
6. Convention centre	1	Cape Town
7. Conservation
a) Nature reserve complexes (nature reserve and residential)	3		Garden Route
			Overberg
			Breede River Valley
b) Conservation and tourist accommodation	2		West Coast
			Overberg
c) Conservation and recreation	2		West Coast
			Overberg
8. Tourist attraction
(a) Themed attraction	1	Blaauwberg
(b) Cultural or e/heritage attraction	5	Cape Town (2)	Winelands
		Tygerberg	Overberg
9. Casino	2	Tygerberg	Overberg
10. Resort	3	South Peninsula	West Coast
			Breede River Valley
11. Waterfront development	1		West Coast
Source: Wesgro, MIDO database.

The most noteworthy and surprising aspect to emerge from Table 5 is the salience of golf-related and property-related or residential developments. Golf-related developments constitute a major component of tourism investments in the Western Cape. By 2001 almost one-quarter (24 per cent) of all tourist developments in the province were golf-related. In total, golf developments (currently under construction and envisaged) carry a value of R7,9 billion, making these projects of the most extensive in the province. A number of the golf projects are in the form of golf courses. The majority of the province's golf projects, however, consist of what are here termed golf complexes – the establishment of golf estates with residences (houses, apartments or villas) or hotels or both. In total there are twelve such developments in the Western Cape. These are mainly countryside developments. The combination of golf, hotel and residential investments – an integrated form of development – is a common and growing trend in real estate development internationally (Miles et al., 2000). International experience shows that two principal motivations underlie such developments from the perspective of investors – land use optimisation and profit maximisation. Close synergies between the three segments in the primary market that is aimed at (a wealthier, golf-playing clientele) enhance the return on investment and hence the attractiveness of such investments (Miles et al., 2000). In addition, diversification reduces the risk of financial loss. With this sort of development, investors in the Western Cape seem to be following an emergent tendency in international leisure development.

The pertinence of the golf developments is significant. It indicates the targeting of a specific niche market – golf tourism. Given the high contingency costs that go along with golf tourism, this market generally consists of a higher-income clientele. Because in the main five-star hotels and middle- to high-income residences form part of these complexes, these can be seen as very exclusive developments. However, a number of golf-related property developments are also what are known as security complexes – security-controlled compounds with limited access points – and this fact suggests that some of these developments are specifically aimed at being exclusionary. This seems a stark contrast to the goals of greater tourism access, empowerment and transformation that underlie the government's policy. In addition, in recent years, there has been increased censure of the negative ecological impacts of golf estates, which generally have high levels of water consumption. As most of these estates are situated in the more rural areas of the province – most new developments have for instance been concentrated in three regions, the Garden Route, Winelands and Overberg – there has been a great degree of criticism of the high levels of water consumption and the added pressures that such developments place on restricted physical resources (Mail and Guardian, 2 April 2003). Notably, these three regions generally draw more higher-income and more international visitors than the other non-metropolitan regions of the Western Cape.

Overall, the high coincidence of residential-cum-commercial and tourism developments indicates a general attempt by investors to optimise investment returns and. as far as possible, lessen investment risks. Very few of the investment projects entail the development of single-modality tourist attractions. Only three of the investments, for instance, consist of the construction or refurbishment of resorts. In addition, only one waterfront development is planned. This development, situated in the coastal town of Lamberts Bay, is part of the West Coast Investment Initiative, one of the Spatial Development Initiatives driven by the central government since the late 1990s. The Spatial Development Initiatives are a comprehensive national strategy aimed at enhancing industrial development in those under-utilised areas with economic potential by removing barriers to investment, establishing infrastructure and providing various financial and tax incentives. The Lamberts Bay project entails the integration of the existing fishing harbour into a waterfront development for tourists.

Several of the investments in the Western Cape are infrastructural development projects. It is noteworthy that very little infrastructure development (with the exception of the centrally driven West Coast Investment Initiative) is taking place in the non-metropolitan parts of the Western Cape. The single largest infrastructure project (both in scale and in financial value) is a medium- to long-term project at the Cape Town International Airport. In the medium term the project entails: the expansion of the cargo and freight centres at the airport; the refurbishment and extension of the runways and the arrivals and departures halls to increase the airport's passenger handling capacity from 350 to 950 passengers per hour; the construction of office blocks and a cricket oval on the airport premises; and the establishment of a 5-star hotel. It is expected that all these developments, valued at R246 million, will be completed by 2005. Longer-term planning, set for completion by 2015 and valued at R2.8 billion, will see a major property development for commercial, retail and warehousing occupation around the airport site, the construction of multi-storey car parks, and satellite terminals and runways (ACSA Master Plan, nd).

In total the value of the entire airport development is R3.1bn, making it the single leading investment project in the Western Cape. This infrastructure project is very important. It is premised on expected rises in passenger and tourist arrivals at Cape Town International Airport. It is projected that traffic at the airport will increase from 5 million in 2000 to more than 12 million by 2015, one-quarter of which will be international tourist arrivals (ACSA Master Plan, nd).

Hotel development is another major constituent of tourism investments in the Western Cape. By 2001 16 individual hotel projects were undertaken or planned in the province, just more than half of this (56 per cent) located in the City of Cape Town. In comparison with the pattern of hotel development in earlier years, there seems to be a greater degree of diversion. Between May 1996 and May 1998, for instance, 32 new hotels were established in the metropolis. Of these, more than three-quarters were located in the Cape Town municipality (Wesgro, 1998). As can be seen in Table 5, by 2001 several new hotel developments were taking place in other parts of the city and the province, suggesting a deconcentration out of the city bowl area. One reason could be that the market in the city bowl has become saturated; Section 4 on the accommodation sector evidenced a large degree of over-capacity in the metropolis.

A second difference is that while earlier hotel developments were mainly mono-functional, several of the new developments that have taken place between 1998 and 2001 are dual-functional or multi-functional. First, as mentioned earlier many of the golf-related or commercial and residential developments are dovetailed with the establishment of hotels. As discussed earlier, this is aimed at offsetting risks. Second, some of the hotel developments are linked with the development of conference facilities. The largest and most important hotel and convention project is the development of the Cape Town International Convention Centre at the foreshore of the city bowl.

A very small proportion of the new accommodation that is being developed is geared towards the segment of the tourist market that does not stay in hotels. Three investment projects, in the Klein Karoo, the West Coast and the Overberg, respectively, consist of the construction of lodges or more informal tourist accommodation. In the latter two regions, moreover, the accommodation developments are linked to broader conservation projects. This indicates that there is only a limited number of large-scale investments aimed at accommodating the middle- to low-income, domestic or lower-paying tourist markets in the province. This may be because these tourists occupy the local guesthouses, backpacker and bed and breakfast establishments in the various parts of the province. A second reason could be that the construction costs of informal or cheaper accommodation establishments such as lodges or chalets are relatively lower, and hence not captured by the MIDO database.

In all, the bulk of large investments target the upper end of the market. Only a small proportion of resources is directed at the domestic, local and lower-paying sections of the tourist market in the Western Cape. In view of the fact that the domestic market is the basis of the tourism industry in the Western Cape, the aptness of this can be questioned.

In sum the following features of tourism investments in the Western Cape may be highlighted. There is a high level of concurrence with property developments in the province (signalling investors' attempt at risk reduction). Coupled with that, there is a broad concentration on multi-modal recreation and leisure complexes. Middle- to high-income tourist markets are generally targeted, but there is a particular targeting of the golf tourist market.

6. Discussion and conclusion: Tourism impact, distribution and development

The following key features emerge with regard to tourism's spatial structure in the Western Cape. Tourist activity is generally concentrated in respect of accommodation supply and usage patterns. There are important differences between regions in the province and municipal areas within the metropolis as regards the markets they draw, their tourist function, and the relative importance of tourism. Tourism is of greatest economic consequence in the Cape Town municipal area, although areas adjacent to the Cape metropolis, most notably the Garden Route and the Winelands, also obtain a substantial proportion of the economic spin-offs of tourism. In the remainder of the province, however, the nature and level of tourism intensity is much less.

Patterns of tourist investment since 1998 have been very particular: the focus is on combination or multi-sphere developments of which tourism is just one aspect. Moreover, demand stimulation is geared at higher-paying, higher-yielding niche markets, rather than the lower-income domestic market. The bulk of investments occur along the coastal areas of the province, and in the Winelands region. These features make the course of tourist development in the province very distinctive: tourist development and value-adding through tourism investment converge in the areas of the province where tourism activity and impact is highest.

To a great extent this is due to the economic and demographic structure of the province, with the greater portion of the population, and the larger component of economic activities based in or adjacent to the metropolis. Yet there are facets to the pattern of tourism development that present significant challenges to the government's goal of fostering a sustainable sector that is of economic benefit for the majority of residents. First, even though there has been wholesale growth in the number of international and domestic tourists to the province, and in tourist activities generally, aspects such as incongruities between supply and demand (such as in the accommodation sector) hold import for tourism growth and sustainability. Declining and seasonal occupation is a generic problem. The hotel sector, however, faces the added difficulties of reduced overall revenues, which in turn influence their employment and other operating practices (this includes the use of casual labour and reduced staff training), and also affects the ability of accommodation operators to engage in aspects such as skills development or training.

Second, while the metropolis is the economic engine of the province, and has accounted for much of the tourism growth over the past decade, this growth has been uneven. A growing body of case studies is showing that tourism benefits are not spreading in the way the government has envisaged. In the City of Cape Town much growth is occurring in the more affluent parts such as the city bowl and Atlantic seaboard, while despite various attempts to promote activities such as township tourism in the more populous and poorer parts of the city, this has not borne the expected growth or benefits (Goudie, Khan & Killian, 1999; City of Cape Town, 2003).

Third, most new capital investments are targeted particularly to the international tourist segment. While the province has a comparative advantage in this market vis-à-vis other South African provinces, to date too little has been done to develop the domestic tourist segment, which has a number of features that should be capitalised on. In the first instance, by its sheer size domestic tourism in the province still accounts for the greater portion of tourist income. In addition, domestic travel patterns do not reflect the excessive seasonal fluctuation of, and are less geographically concentrated than, the international market. For example, although there is a general slump in domestic visitor numbers over the winter period (May to August), areas such as the Breede River Valley and the Central and Klein Karoo experience relatively high levels of tourism; for instance, 40 per cent of all visits to these areas are made in the winter (WCTB, 1999). These patterns suggest a greater degree of consistency than that showed by the international segment.

The provincial government acknowledges that domestic tourism is important and should be promoted (DEATT, 2001). It has pursued this objective through a series of product development and promotion strategies, of which the aim to ‘package tourism themes and attractions in the form of suggested routes and itineraries’ (DEATT, 2001: 38), and the development and promotion of new, alternative products such as cultural tourism, are the most significant. The latter aims to integrate people previously excluded from tourism into its production, while with the routes and themes concept the goal is to provide more exposure to the hinterland of the province.

The province shares similarities with many other developing countries in that its tourism development policy is a highly ambitious attempt to merge two very different, often contending objectives – sustainable growth, and equity or distribution. This means satisfying both the needs of the market and the requirements for distribution.

Tourism literature suggests that both objectives – sustainability and empowerment – are in practice difficult to realise. In other developing countries this is due to factors such as insufficient infrastructure and resources, lack of capacity (Goudie et al., 1999), poverty (Sindiga, 1996), low levels of education and weak institutions. South African studies have noted how the shortage of skills, capital and other resources has posed obstacles to community-focused tourism programmes (Koch, 1994; Boonzaaier, 1996). In this study, interviews with a wide range of tourism producers, regulatory and organisations and tourism officials in the Western Cape indicated that the province is similarly hampered by inadequate resources, along with, a general lack of interest in township or cultural tourism by the international market. In addition, commercial considerations often mean that large, well-established tourism companies are unwilling to contribute to empowerment or transformation goals.

The skewed spatial structure of tourism in the Western Cape indicates that there is much potential to distribute tourism activity and impact. Provincial policy, specifically as encapsulated in the 2001 White Paper, has appropriate goals, but to date has failed in implementation. More attention should be given to addressing the institutional shortcomings of tourism administration, which despite attempts to streamline the various regional and local institutions, still remains too bulky. The establishment in early 2004 of the Destination Marketing Organisation would hopefully go some way to improving tourism governance. At the same time, concerted effort needs to be made to address lack of capacity at the local tourism institutional level, particularly in the poorer parts of the city. In addition product development should be in line with consumer's preferences and there should be a general attempt to increase the domestic market. In this, the government has some role to play in channelling capital investments that aim to promote domestic tourism. Providing incentives to the private sector would be an important part of this.

Notes

1 Tourism Satellite Account (TSA) methodology was developed by the World Tourism Organisation (WTO) and the World Travel and Tourism Council (WTTC). It aims to provide a comprehensive delineation of all the economic activities related to, or influenced by travel and tourism in a given economy, by anticipating and depicting the flow-through effects of tourism-specific and tourism-related activities.

2 One example is of multiplier models (see, for instance, Archer, 1977) that attempt to measure the direct, indirect and induced effects of tourist expenditure on income. Another, related methodology is input-output analysis that attempts to measure the effect and flow-through of tourist or government spending, or tourism-related investments (Fletcher, 1989; Fletcher & Archer, 1991). A third is Tourism Satellite Account (TSA) methodology.

3 This was part of a larger analysis of provincial tourism in which a total of 181 interviews were conducted with local and international tourism producers, regulators and consumers.

4 The director of one of the largest hotel chains in South Africa, for instance, argued: ‘The situation in 1986 was such that we had a state of emergency, we had disinvestment, sanctions. That was the lowest period in apartheid-era tourism, and this affected hotels severely. Occupancy in our hotels, in our industry, was at an all-time low, there were no new projects happening whatsoever. Hotels were unable to refurbish because they didn't have the revenues. Then, after the political transformation in 1994, things changed dramatically. We ended up having dramatic growth after 1996.’

5 Section 13.2 of the Income Tax Act of 1967 offers incentives for the construction and development of new hotels. Up until 1988 tax concessions were given for the development of three- to five-star hotels. Such concessions included the rescinding of building costs over 9-, 12- and 14-year periods. In 1988 20-year write-offs for building costs were introduced for hotels of all star gradings (DEAT, 1996).

6 This is due to the demographics of casino patrons, who are generally medium- to low-income local residents or day visitors from further afield.

7 Personal communication with owner and chief executive officer, Relais Hotels, 6 March 2001.

8 Data obtained from Statistics South Africa (Stats SA).