Special Issue: The Economics of Cultural Tourism – New Topics and Methods Festival value in multicultural contexts: City festivals in South Africa Tourism Economics 1–20 ª The Author(s) 2020 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/1354816620932808 journals.sagepub.com/home/teu Jen D Snowball Rhodes University, South Africa; South African Cultural Observatory, South Africa Geoff G Antrobus Rhodes University, South Africa; South African Cultural Observatory, South Africa Abstract Worldwide, the number and variety of cultural festivals have grown dramatically. Many areas see festivals as an important way to attract tourists, and their spending, to a region, resulting in a positive economic impact. While they offer important opportunities for artistic producers and audiences, there is growing pressure for festival organizers to demonstrate their value to society beyond their economic impact. Like many countries, South Africa has a strong focus on increasing diverse cultural participation, demonstrating the social, nonmarket values of events that receive public funding. Using data from two South African festivals, the article uses a valuation framework developed by the South African Cultural Observatory to demonstrate measures of audience diversity, the use of quality of life measures to gauge the impact of culture on well-being, and the use of community focus groups to assess the impact of participation on social cohesion and capacity building. Keywords diversity, evaluation, festivals, methods Introduction In the past 30 years, many local governments have promoted cultural tourism as a local development strategy, with the focus on the new money that tourists bring to a region and the increased incomes and jobs that these generate. Bonet (2011) attributes the rise of cultural tourism to a number of factors, including the growing middle class and the diversification of Corresponding author: Jen D Snowball, Department of Economics, Rhodes University, PO Box 94, Grahamstown 6; 140, South Africa. Email: email@example.com 2 Tourism Economics XX(X) tourism and travel. Cultural tourists travel away from their normal residences for the purpose of experiencing and learning new things related to cultural, symbolic, or spiritual values. In South Africa, tourism has long been promoted as a local economic development strategy (Rogerson, 2008) and Toerien (2020) shows that tourism in smaller towns is associated with lower levels of community poverty. Worldwide, the number and variety of cultural festivals have grown dramatically. Cultural festivals can play an important part in attracting tourists to a city, as well improving the city’s image and thus encouraging inward investment. For local residents, cultural events can promote the development of social capital, communal solidarity and identity, and leverage the development of infrastructure (Quinn, 2010). There are currently more than 600 festivals per year in South Africa, many of which are experiencing a drop in visitor numbers (and spending) as household incomes decline in a climate of slow economic growth (Van Heerden and Saayman, 2018). As in other countries, many of these festivals also receive public support and are facing pressure to demonstrate their value to society as public goods, as well as through economic impact (Ferguson, 2013; Herrero et al., 2001). In postapartheid South Africa, demonstrating how festivals are expanding participation and audience diversity is a key factor. In such a climate, it is no longer enough to provide evidence of the financial impact of an event through increased tourism—a more nuanced, multidimensional valuation framework is needed. In a valuation, the characteristics of the host city itself can make a big difference to both the social and economic impact of the event. The socioeconomic conditions may also influence the aims of a particular event, and the structure and size of the local economy affect the way spending is translated into economic impact. Cultural festivals that take place in big cities have particular challenges and opportunities compared to those held in smaller towns. For example, cities can provide larger local audiences and easier access to specialist services, artists, and performers. On the other hand, in cities, many residents and visitors may not even be aware of the event taking place, making it more challenging to demonstrate local impacts. In 2016, the South African Cultural Observatory (SACO) produced the Framework for the Monitoring and Evaluation of Publically Funded Arts, Culture and Heritage. The Framework provided a method for valuing a wide variety of cultural events, projects, and institutions under five cultural value “themes,” which were then linked to indicators, which included both quantitative and qualitative measures. As part of testing the framework, it was applied to a wide range of cultural festivals in South Africa, two of which this article discusses: The Cape Town Carnival and the Mangaung African Cultural Festival (MACUFE), both of which take place in two of the larger South African cities (Cape Town and Mangaung/Bloemfontein). Both the host cities, and the festivals themselves, have very different characteristics and aims. The goals of this article are to demonstrate the importance of using a variety of value indicators when assessing the impact of cultural festivals, especially in multicultural societies, like South Africa. The article places particular emphasis on indicators of audience diversity, well-being, and participation, in addition to the more usual economic impact methods. Results showed that, in some circumstances, festivals are much more successful in attracting tourists from outside the host city and thus generating economic impacts. However, this needs to be balanced by considering the noneconomic, social impacts of other kinds of cultural festivals, which may be more important to local residents. Too heavy an emphasis on tourist and economic impact may reduce funding to smaller events which have equally important cultural and social benefits. Snowball and Antrobus 3 Festival value and the SACO framework The values associated with arts, culture, and heritage can be divided into three broad categories: economic (financial) impacts, social impacts, and the intrinsic value of art itself. Nadotti and Vannoni (2019: 117) argue that “A chain of relations between culture and tourism could be ideally defined around the concept of an event,” since the event is a manifestation of cultural and social values, while tourism links to economic and placemaking goals. Cultural festivals thus provide an interesting opportunity to develop and apply multidimensional measures of value. Economic impacts Economic, or financial, impacts come about as a result of the inflow of new money into an economic system as a result of visitors from outside the region. Visitors spend money on accommodation, transport, food, shopping, tickets, and so on. This spending then recirculates in the host economy, thus increasing sales and employment in local businesses. Festival evaluations tended to be dominated by economic impact methodology to the exclusion of other important measures of value (Quinn, 2016), despite the increasing skepticism about the validity of such studies (Nadotti and Vannoni, 2019). Social values Social values relate to the benefits to society, such as education, creativity, and innovation, social cohesion, and identity formation (Bohm and Land, 2008). Cultural festivals can be important sites of interaction for both producers and consumers. For producers, Ferguson (2013) describes arts festivals as “touchstone experiences” that contribute toward the development of networks and social capital. Festivals also give artists access to peer-to-peer pedagogy and “inspirational capital” which contribute to the development of symbolic capital (having an influence in the field) and increase the probability of artistic, and often financial, success (Ferguson, 2013). For audiences, festivals offer opportunities to experience new, or a wider variety of, cultural forms or genres, as well as the social aspects of spending time with family and friends, both of which could increase quality of life (QoL) (Hand, 2018). Jordan (2013: 11) conceptualizes festivals as spaces that promote experimentation for both producers and audiences: “The other worldliness of the festival environment provides a safe space in which to try out different personas or activities that can either be incorporated into or be rejected on return to everyday life.” Intrinsic value Intrinsic values are related to the symbolic and artistic nature of the product itself and to feelings invoked in individual participants (such as joy, sadness, anger, delight, questioning, etc.). Development of cultural indicators themselves starts with the identification of the general categories of cultural value of most interest. Rather than beginning with the more easily quantifiable economic, or even social indicators, the intrinsic starts with “people’s experience of arts and culture, rather than secondary benefits that follow” (Crossick and Kaszynska, 2014: 123). Measuring intrinsic values may need to incorporate qualitative data and theory from disciplines other than economics. 4 Tourism Economics XX(X) Cultural participation, diversity, and well-being With the increasing importance of accountability for public and private sponsorship, data on participation rates and on who participates in cultural events are also becoming more important and contentious. Ideas of what cultural participation means are intimately linked to the cultural policy paradigm of the country (Bonet and Negrier, 2018). For example, in a paradigm of “Cultural Excellence,” passive audiences require high levels of cultural capital to appreciate “high” art. In this scenario, a lack of participation is conceived as being a result of a “deficit” on the part of the audience (their inability to appreciate “excellent” art). With the recent shift in cultural paradigm toward “Cultural Democratization” (Bonet and Negrier, 2018), cultural policy emphasis changed from supply-side issues of excellence to demand-side focus on audience access. “Cultural Democracy” takes this even further, conceptualizing participation in a much more active way through, for example, amateur production and crowdfunding. This links to the theory of “cultural ecosystems,” which include not only the formal, professional production of arts and culture but also the underpinning informal and amateur activity that supports the final product sold in the market (Wilson et al., 2017). Many studies have shown that cultural participation is determined by socioeconomic class, including family background and education as well as household income (Mulcahy, 2017; Throsby, 2010; Van Hek and Kraaykamp, 2013; Willekens and Lievens, 2016), which seem to support the idea that the arts are consumed or appreciated mostly by higher income and educated members of society. Glow et al. (2020: 1) note that audiences in Anglophone countries remain predominantly “white, middle class, and middle-aged.” This is despite a strong cultural policy focus on diversification in many countries, and the importance of inclusion both in terms of social justice and equity, as well as for the sustainability of arts organizations themselves (Werner et al., 2014). In South Africa, with its apartheid history and ongoing socioeconomic inequality, improving the diversity of audiences is a key goal of the Department of Arts and Culture and a requirement of many public and private funding agencies. Audience diversity in relation to cultural events can be defined in terms of many indicators, including household income, gender, age group, race, language, ability, education levels, and motive for attending (Glow et al., 2020; Werner et al., 2014). Research has shown that improving audience diversity requires long-term organizational change and commitment (Glow et al., 2020) and that focusing on a few limited demographic descriptors, like race and ethnicity, are of limited use in understanding, and thus reaching, diverse audiences. There is also increasing interest in measuring cultural value through an analysis of the relationship between QoL, also referred to as subjective well-being, or happiness, and cultural participation (Galloway, 2006; Hand, 2018; Michalos and Kahlke, 2008; Steiner et al., 2015). Research in South Africa (Botha and Snowball, 2015; Ebrahim et al., 2013) and in other parts of the world has found that QoL is affected by many factors. Some of the most common are age, gender, income, education, employment, health, personal relationships (especially with life partners and children), relative social status, and control over one’s life. There is also some evidence that consumption of “experience” goods, such as attending a festival or travel, increases life satisfaction more than purchasing goods. However, despite much qualitative research into the relationship between QoL and cultural participation or consumption, there have been relatively few quantitative studies in this area. A problem with econometric QoL studies and culture is that, while there may be positive correlations between cultural consumption and subjective well-being, other factors often have a larger effect Snowball and Antrobus 5 and are also highly correlated with culture (e.g. education and cultural consumption often go together). This makes it difficult to isolate the size of the impact of culture on QoL (Galloway, 2006). A recent article by Hand (2018) used a large database and an econometric technique that allowed arts attendance and other control variables to vary across different levels of happiness. His findings showed that, especially for those with lower levels of general life satisfaction, cultural consumption did statistically significantly increase QoL. The SACO framework Based on the three broad categories of value discussed above (economic, social, and intrinsic), the Framework for the Monitoring and Evaluation of Publically Funded Arts, Culture and Heritage (SACO, 2016) was developed. The choice of themes was based on a review of local and international literature on cultural value, some of the key goals for South African cultural policy. Specific areas of importance for one of the main public cultural festival funding sources from South Africa’s national Department of Arts and Culture were also considered. To operationalize the framework, the themes were linked to indicators so that data or evidence about the impact of the specific project or organization could be presented (see Online Appendix 1). The idea of the SACO framework was not that all cultural festivals report on all themes and indicators, but that they select those that are most applicable to the aims and values of the event itself. There are a wide variety of frameworks for evaluating the impact of culture. The UK “Cultural Value Project,” for example, took an open-ended, multidisciplinary approach, recognizing, as Holden (2009) does, that the indicators and methods chosen should depend on why a particular valuation is taking place. Similarly, Crossick and Kaszynska (2014: 128) question the “belief in a single quantifiable measurement to articulate the importance of arts and culture relative to other activities” and argue instead for a much broader approach in shaping a research agenda for cultural value. The successful attraction of cultural tourists to a festival depends on effective event management, which includes the systematic monitoring of the needs and experiences of tourists (Nadotti and Vannoni, 2019). Equally however, too strong a focus on the instrumental values associated with a cultural event (attracting tourists spending) will result in a loss of its inherent cultural meaning, which will reduce the degree of authenticity of a tourist’s experience, and impact negatively on the longterm sustainability of both the event and the tourism it attracts (Ivanovic, 2008). The SACO framework outlines of five broad cultural value themes: audience development and education; human capital and professional capacity building; inclusive economic growth; social cohesion and community development; and reflective and engaged citizens. An explicit goal of the framework was to make a contribution to the valuation of cultural festivals that places emphasis on values in addition to the economic and that was applicable to a wide range of different kinds of cultural events. For the purposes of this article, a few key themes and indicators that are particularly applicable to the aims of the cultural festivals being studied, and that made use of both quantitative and qualitative data, are discussed. Background and context: The Cape Town Carnival and MACUFE The Cape Town Carnival The City of Cape Town Metropolitan Municipality is South Africa’s second largest metro by population size (3.7 million). Despite a relatively diverse economy, it has an unemployment rate of nearly 24%, which is nevertheless lower than the South African average. The largest language 6 Tourism Economics XX(X) group is Afrikaans (35%), followed by isiXhosa (29%) and English (28%). Compared to many other parts of South Africa, living conditions are good, with 78% of the population living in formal housing and 94% having access to electricity for lighting. Cape Town is well-known as a tourism destination for both local and international visitors, especially for its scenic beauty and for the wine industry in the near vicinity. According to a recent tourism report (Statistics South Africa, 2015), 20% of international air travelers enter the country through the Cape Town International Airport, which amounts to more than half a million international visitors per year. Although the Cape Town metropolitan area is regarded as one of the affluent and developed parts of South Africa, there are large areas consisting of informal housing, or shacks, mostly occupied by black and “colored” South Africans. This is despite large-scale housing projects that have been in place since the end of apartheid in 1994. The Cape Flats area on the outskirts of the city is also known for its gang violence and poverty. The Cape Town Carnival was established in 2007 and is operated by a nonprofit trust. The Carnival Parade is free of charge, although a limited number of seated tickets on stands were provided at recent Carnivals. The Carnival has grown significantly since 2010, with around two dozen floats, an estimated 50,000 spectators, 1900 performers, and 50 participating groups (Cape Town Carnival Annual Review, 2016). The Carnival—to the public—is a 1-day event. Most of the core objectives of the Carnival, however, are achieved in the months leading up to the public event during which float builders, musicians, artists, costume designers, performers, dancers, school, and other groups prepare for the parade. The Carnival Parade itself is positioned as a “glamourous celebration of African identity, of diverse communities and the transformative power of creativity” (Cape Town Carnival Annual Review, 2016: 6). The aims of the Carnival, including artistic, economic, and social goals are, staging a world-class parade that is safe and well attended; providing opportunities for community participation in areas and schools that previously would not have had a purpose or incentive to do so; enhancing networks and collaboration in the creative community, building skills, job creation, and training; and building social cohesion. In addition, the Carnival hopes to create a hub for clothing, costume, and set-building industries in the Western Cape, which is complementary to the already established film industry. The MACUFE The Mangaung Metropolitan Municipality is located in landlocked Free State province and is the eighth largest municipality in South Africa. The city of Bloemfontein, where MACUFE takes place, is the provincial capital and has a population of nearly half a million people. The majority of people living in the municipality are Black Africans (83%), with the largest language groups being Sesotho (52%), Afrikaans (16%), and Setswana (12%). The unemployment rate is nearly 28%, which is four percentage points higher than in Cape Town, and the youth unemployment rate is 37%. However, 84% of people live in formal dwellings and 91% have access to electricity for lighting (Statistics South Africa, Census, 2011). The main economic activities in the Free State province are agriculture and mining, neither of which have recently grown significantly, according to the Mangaung Metropolitan Municipality Integrated Development Plan Report (2017). Tourism is mostly domestic and occurs around specific events, such as festivals and sports. Tourism development strategies are generally around Snowball and Antrobus 7 the development of various routes (e.g. Heritage route; Tolkien trail route, Batho route), which seek to link smaller heritage sites in the area into more attractive packages. MACUFE takes place annually over 10 days in late September to early October. The festival includes art, music, theatre, craft, and sports exhibitions throughout the city. MACUFE has been running since 1997, making it one of South Africa’s most established arts and culture events. The vision of the Festival is “to be the biggest, most culturally balanced showcase of African Arts and Culture in the world.” The main events of the Festival are focused around music, including mostly South African artists, but also some high-profile international performers from the United States and other African countries. However, it also includes stand-up comedy and an arts and crafts market, which offers live entertainment, a beer garden, and a Basotho Cultural Village. Entrance to this part of the Festival is free. Some venues are outdoors, or in specially erected marquees. By South African standards, ticket prices are relatively high, ranging from R200–R350 (US$14–24). MACUFE also incorporates a smaller theatre component through “Development Performances, Local Theatre, Drama and Poetry,” and an “Urban Youth Festival” music event and party, which have much lower ticket prices, and also offers sports events, such as a boxing tournament and a soccer match (Sparta MACUFE Cup). The Festival is organized by a private events management firm on behalf of the provincial Department of Sport, Arts, Culture and Recreation. In 2013, the additional funding required from various government sponsors to cover the costs of the event sparked some controversy (Free State Times, November 23, 2013). However, the 2015 Festival received some good press coverage emphasizing the growing international profile of the event as including international artists, and attracting audiences from Lesotho, Namibia, and Mozambique. Values that were emphasized included providing emerging South African artists with exposure to promoters, expanding tourism in the province, and economic impact. MACUFE thus takes place in a rather different environment from the Cape Town Carnival. The city of Bloemfontein is not known for tourism and is focused mostly on servicing the agricultural and mining interests of the province. In addition, it is a much smaller city than Cape Town and less culturally diverse. Issues of social cohesion, broadening participation, and building interracial understanding are thus not so prominent at MACUFE. Rather, the challenge is to establish the city as a viable tourism destination for nonlocal residents. Research methods At both festivals, primary data were collected using previously tested survey instruments and trained enumerators. At the Cape Town Carnival, 396 interviews were conducted with attendees on the day of the Carnival. Since the event is primarily a street festival, no stratification of the sample was possible, beyond instructing interviewers to include as wide a variety of people as possible, so random sampling was used. Interviews were conducted throughout the event, from early in the day to late in the evening. The questionnaire collected information on audience demographics (age, race, home language, and education level), home town, income, length of stay, and opinions about the event. Event organizers provided information about spending and sponsorship. Carnival performers include local, community-based organizations, who work with professional performers and designers. An important aim of the Carnival is to build social cohesion and community identity through the event, as well as to develop skills in the cultural and creative 8 Tourism Economics XX(X) sectors. After the Carnival, two focus groups were held with 20 participants who represented 18 of the local community groups involved. The open-ended questions included a discussion of the extent to which the Carnival represented the cultural diversity of the city and was successful in fostering social cohesion; and whether the Carnival did provide opportunities for skills development in the cultural sector. At MACUFE, interviews were conducted at all the major events during the Festival, weighted by the expected audience size. While both local residents and visitors from outside the impact area were interviewed, the focus was on visitors. An earlier study by the University of the Free State (2015) Centre for Social Development tracked the number of people interviewers had to approach before they found a visitor (tourists) as a means of estimating the proportion of local residents compared to visitors at the various events. It ranged from an average of 1.7 at the larger events to 3.5 at smaller events. Using this as a guide, a quota of 20% for local residents and 80% for visitors was set. In all, 553 interviews were conducted (124 with local residents and 429 with visitors). Information was collected on respondent demographics (gender, home language, and age), place of residence, festival activities (free and ticketed events), spending, and length of stay, as well as opinions about their festival experiences. The MACUFE study also included a subjective QoL measure (discussed further below). Post-event, organizers provided data on their spending, sponsorship received, ticket sales, and festival objectives. Methods of analysis included quantitative methods (such as the production of summary statistics, regression analysis for the QoL analysis, and economic impact analysis based on regional multipliers) and qualitative methods (such as thematic analysis). Results and discussion Measuring tourism and audience diversity Understanding the origin of visitors is important for cultural festivals, especially if one of the goals is attracting cultural tourists as a way of generating economic impact and becoming recognized nationally or internationally. For festivals with a mandate to include and benefit local residents, the proportion of those from the host city may be an important indicator of regional nonmarket value. However, determining the number of visitors and whether the research has drawn a representative sample can be a challenge for festivals with no tickets or defined entry points. For ticketed events, a potential source of information is where the tickets were purchased: one can assume that if the tickets were bought outside of the host city, at least this proportion of visitors was nonlocal. For example, data on where tickets for the largest four shows at MACUFE were bought showed that 32% of tickets were sold in the host city Bloemfontein/Mangaung, while 68% were sold outside the impact area before the festival took place. Since tickets could also be purchased during the festival in the host city, this finding supports the survey data which showed that the majority of festivalgoers (78%) were tourists, not local residents. Almost all the tourists interviewed reported that MACUFE was their main or only reason for being in the city. In terms of visitor origins, 40% came from the province in which the festival was hosted and 3% were from outside of South Africa. In contrast, the Cape Town Carnival survey showed that 87% of attendees were local residents. Of the tourists that the event attracted, only 41% reported that the Carnival was their main or only reason for visiting Cape Town; 30% of tourists were from the province in which the even was hosted; and 6% were from countries other than South Africa (including Australia, Canada, China, the United Kingdom, Germany, Korea, Saudi Arabia, the United States, and Zimbabwe). Snowball and Antrobus 9 The results illustrate that both the characteristics of the host cities, as well as the type of cultural festival hosted, impact on their ability to attract cultural tourists. Cape Town is a larger city than Bloemfontein/Mangaung, resulting in a higher proportion of local residents attending the event. Although widely advertised, the Carnival focus is particularly on the involvement of local residents and the celebration of diverse local cultures, both of which characteristics are likely to have increased the proportion of local festivalgoers. The higher percentage of foreign visitors is indicative of the international tourist appeal of Cape Town, which was not apparent in Bloemfontein/ Mangaung. Another important figure for determining economic impact, as well as the importance of the event in place marketing, is the proportion of visitors who were in the city specifically to attend the festival, rather than those that happened to be present (referred to as “casuals”), or were coming to the city for something else and changed the time of their visit to coincide with the festival (“time switchers”). As shown (Table 1), results varied considerably between the Carnival and MACUFE: Nearly all nonlocal visitors who attended MACUFE were there specifically to attend the festival (94%), while less than half of the nonlocal visitors who attended the Carnival were in Cape Town mainly or only to do so (41%). These differences can also be explained by the characteristics of the host cities and the events: Bloemfontein/Mangaung does not offer many other tourist activities or attractions, making it unlikely that visitors to the festival would be time switchers or casuals. Cape Town, on the other hand, is a well-established tourist destination, offering many other attractions. MACUFE mostly consists of formal ticketed events, performed by professionals, with one of the main aims of the event to raise the profile of the city through formal cultural offerings. This conceptualization of “participation” thus fits more with the cultural policy paradigm of “Cultural Excellence” or “Cultural Democratization,” where the emphasis is on the high quality of artistic performance in particular genres (albeit popular ones), with audiences “consuming” a professionally produced “product” through the purchase of a ticket. The Carnival consists of both professional and amateur performers, with one of the main aims being to increase active participation of a diverse range of local residents, including local schools and community groups who participate in the construction of the floats. In this instance, participation fits with the “Cultural Democracy” paradigm, where the lines between audience and performer are blurred, and a much more active mode of participation is envisaged. There were also significant differences between the two events in terms of audience diversity (Table 2). Both festivals attracted about equal proportions of men and women, but the Cape Town Carnival audiences tended to be younger, perhaps as a result of the less formal setting. In terms of race groups, more than half the spectators were “colored” (57%), followed by Black Africans (21%), 16% White people, and 4% Asian or Indian-origin people. Overall, 5% of festivalgoers who were interviewed were disabled or had in their group someone with a disability. An important way of assessing diversity is an examination of the various cultural groups represented by the audience, rather than simply focusing on racial classifications as is often done in South Africa and elsewhere. The demographics of the MACUFE audience appear, at first glance (Table 2), to be much less diverse than the Carnival audience, since the vast majority of MACUFE attendees (96.3%) were Black Africans, with much smaller groups of White (0.6%), colored (2.9%), and Indian/Asian-origin people (0.1%). However, more cultural diversity is found within home language groups at MACUFE than at the Cape Town Carnival. Although nearly half (46.1%) of MACUFE respondents spoke Sesotho at home, about a fifth (19.7%) spoke isiXhosa, with smaller groups speaking Setswana, isiZulu, and Afrikaans. The “Other” 10 Tourism Economics XX(X) Table 1. Visitor origins at the Cape Town Carnival and MACUFE. Category Local residents of host city as a percentage of festivalgoers Non-South African festivalgoers (%) Nonlocal festivalgoers from host city province (%) Nonlocal festivalgoers whose main or only reason for being in the city was to attend the festival (%) Cape Town Carnival MACUFE 87 6 30 41 22 3 40 94 Note: MACUFE: Mangaung African Cultural Festival. Table 2. Audience diversity at the Cape Town Carnival and MACUFE. CTC categories Gender Age groups Population groups Home language Education Household income CTC (%) Women Men Under 20 20–35 36–50 >50 Black African Colored White Indian/Asian English Afrikaans isiXhosa Other South African Other 46 54 25 40 23 12 21 57 16 4 66 15 9 3 7 School not completed High school completed Professional qualification/ degree Up to R10,000 R10,001–R30,000 R30,001 or more 24 37 39 44 40 17 MACUFE categories Women Men 18–25 26–35 36–50 >50 Black African Colored White Indian/Asian Sesotho isiXhosa Setswana isiZulu Afrikaans Other School not completed High school completed Professional qualification/ degree Up to R10,000 R10,001–R30,000 R30,001 or more MACUFE (%) 50 5 16 43 34 8 96 3 1 <1 46 20 16 5 3 10 1 29 70 31 38 31 Note: MACUFE: Mangaung African Cultural Festival. category included a small number of people (15 or less in each group) who spoke Sepedi, Ndebele, Shona, SiSwati, Xitsonga, and Tshivenda. By contrast, two-thirds (66%) of Cape Town Carnival festivalgoers spoke English as their home language. The next biggest language group was Afrikaans (15%), followed by isiXhosa (9%) and other South African languages, and languages from other countries. Demographic indicators of audience diversity thus need to be multifaceted and context-specific. Snowball and Antrobus 11 Table 3. Regression results for MACUFE quality of life indicator. Variable Constant Age group (categorical) Education level (categorical) Employment status (1 ¼ employed) Gender (men ¼1) Like African Music Score (1–5 scale) Attended other music festivals in the last year (1 ¼ yes) Adjusted R2 F-statistic N Coefficient (standard error) 3.038033*** (0.312511) 0.063713 (0.061846) 0.028695 (0.053838) 0.214489* (0.123197) 0.004886 (0.096027) 0.193384*** (0.060272) 0.068193 (0.104116) 0.03 3.326*** 501 Note: MACUFE: Mangaung African Cultural Festival. *Significant at the 10% level. ***Significant at the 1% level. Income and education groups can also be important socioeconomic indicators of cultural participation. At MACUFE, audiences were generally better educated (70% of participants had a professional qualification or degree), and a greater proportion reported having household income in higher categories (31% had household income for more than R30,000 per month). At the Cape Town Carnival, much lower proportions of participants reported having tertiary education (39%) and a much higher proportion were in the lower household income categories. It is clear that the more commercial MACUFE, with higher ticket prices and the emphasis on professional performers, attracts a different demographic from the Carnival. What the discussion of audience shows is that the context and type of event is important in assessing diversity. MACUFE is a much longer event, with formal ticketed shows. The majority of attendees were both higher income and education visitors from outside the city, which increases the economic impact of the event because of visitor spending—an important goal of the festival. Although MACUFE audiences are not racially diverse, they are culturally diverse in terms of home language groups. The Cape Town Carnival, on the other hand, has a much shorter time frame and attracts mostly local participants, who are racially diverse and some of whom come from poorer communities. Although thus not as successful as MACUFE in attracting tourists, the Carnival audience fits with the community and social cohesion aims of the event. QoL measures: A quantitative analysis The MACUFE study included a question asking respondents to rate their life satisfaction from 1 to 5, where 1 meant totally unhappy and dissatisfied and 5 meant totally happy and satisfied. The results were then correlated with other variables, after which an ordinary least squares (OLS) regression was run to determine their statistical significance when holding variables constant (Table 3). More than half of respondents at MACUFE rated their life satisfaction as 5 (totally satisfied), with a further 20% rating it very highly (4 out of 5). Only 23% rated their life satisfaction as a 3 or lower. The average score was 4.20.1 Although MACUFE is a festival that covers many cultural genres, the majority of events are related in some way to music (jazz, gospel, divas, and “main” 12 Bridging funds made available; pre-Carnival workshops Facilitating valuable for skills and network development access and collaboration Note: CTC: Cape Town Carnival. Development of professional capacity Inclusion Cultural diversity and social cohesion Positive dimensions: What worked well for participants The CTC team reaches out to all the communities regardless of location or economic status; making sure that everybody works together as one; identification of broad, inclusive Carnival theme Participation Meetings where group leaders get to know each other and develop healthy relationships; interaction with people of all cultures and ages makes it easy to make a connection Recognition The CTC reaches out to communities that do not have a way of being seen or noticed Leadership and Managed by an extremely talented and creative team who coordination create a family environment. High levels of trust between participants and Carnival management team Training and Working with professionals; informal interactions where experience knowledge, skills and networks are shared Subtheme Theme Table 4. Qualitative analysis of community focus groups at the CTC. Need greater continuity and more events throughout the year; greater inclusion of groups from rural areas; use social networks to stay in touch Take the Carnival to the communities from which groups come; facilitate greater media coverage Organize a formal debrief to avoid anticlimax; explore ways of overcoming racial and locational definitions of “community” Better leveraging of opportunities for participants to join the creative economy (start-ups; intergenerational training; use of skills already in the group) Provision of transport; better use of social media to enable greater interaction between participants; developing and sharing a database of participant skills After-party is not for everyone; conduct a survey of different cultures in Cape Town and invite them all Negative dimensions: Aspects that need improvement Snowball and Antrobus 13 Table 5. The economic impact of MACUFE and the Cape Town Carnival (in millions of Rands). Category Total organizer amount spent locally (in millions of Rands) Net visitor spending (in millions of Rands) Total captured spending in impact area (in millions of Rands) Indirect multiplier effects: Local output multiplier Total economic impact (in millions of Rands) Cape Town Carnival MACUFE 8.8 13.8 22.6 1.82 41.1 26.0 36.6 62.7 1.65 103.4 Note: MACUFE: Mangaung African Cultural Festival. music festival), particularly celebrating African music and musicians. Other than the soccer, these are also the events which attracted the largest audiences. To reduce the number of variables in the study, the focus of this exploratory research was specifically related to the relationship between African music and QoL. To this end, respondents were also asked to rate their enjoyment of African music on a 1 to 5 scale. Building on the theory of “rational addiction” in cultural consumption (i.e. that increased consumption increases intensity of appreciation, which in turn increases consumption), it is reasonable to assume that those with a greater intensity of liking for African music are more likely to be frequent consumers. However, in a developing country context where many people have low levels of disposable income to spend on attendance at cultural events, appreciation of music may not necessarily correlate with attendance. Unsurprisingly, given that the interviews were being conducted at a festival with a large African music component, the vast majority of MACUFE respondents rated their enjoyment of African music as either a 4 (30.4%) or 5 (56.7%) out of 5. The average score was 4.4. To determine other live music consumption, respondents were also asked if they had attended “any other live music events or festivals in the last year, other than MACUFE.” Two-thirds (67%) of respondents said that they had done so. An OLS regression was then used to investigate the relationship between QoL, the self-reported enjoyment level for African music, and the frequency of attendance at other cultural festivals, by MACUFE audiences. Control variables were sex, age group, education level, household income, and employment. However, given the high correlation between household income and education (0.4), the regression analysis excluded household income. The following model was run: QoLi ¼ C þ aAgei þ bEduci þ gEmpli þ dSexi þ qLikei þ tOtherFesti þ ei : The model did not perform particularly well, explaining only 3% of the variation in life satisfaction. Of the control variables, the only statistically significant one (at the 10% level) was employment, indicating that being employed had a positive impact on QoL score, while holding all other variables constant. This is supported by a previous finding by Ebrahim et al. (2013), which showed that, in South Africa, employment status has a bigger impact on life satisfaction than earnings. The analysis did show a strong positive relationship (at the 1% level of significance) between the liking for African music score and life satisfaction. Holding all the other variables constant, a one-unit increase in liking African music score is associated with a 0.19-unit increase in life satisfaction. Interestingly, the frequency of attendance at other cultural festivals over the preceding 3 years was not a statistically significant determinant of QoL. 14 Tourism Economics XX(X) The data from MACUFE suggest that a greater liking for the kind of music offered at the festival is positively related to higher levels of personal well-being or QoL for both tourists and local residents. This is an important finding in terms of the “intrinsic” cultural value theme. Such values are related to individual emotion and spiritual responses to arts and culture and linked to the notion of artistic quality or excellence (Holden and Balta, 2012). These nonmarket values are often difficult or impossible to measure in monetary terms, but may be just as, if not more important than, instrumental values in demonstrating the importance of the arts and in justifying public support (Snowball, 2011). As one of the most important African cultural music festivals in the country, MACUFE plays a significant role in increasing cultural supply, both through hosting the festival, and by giving African artists exposure and experience that may help to develop their careers. By increasing the supply of African music, MACUFE plays a part in increasing the QoL for both local residents and tourists. Some limitations of the study need to be acknowledged. Data were collected at a specific cultural festival with a strong focus on African music and artists. Since interviews were mostly conducted at ticketed events, respondents were also mostly limited to those who could afford to purchase a ticket. It should also be noted that the question was about the intensity of liking for African music, rather than actual attendance or participation at live music events. More work needs to be done before results can be generalized to other groups. Cultural participation and social cohesion The Cape Town Carnival blurs the line between consumption and production in that one of their very important goals is to build social cohesions through community participation. In terms of the cultural policy paradigms discussed earlier, their approach could be described as “Cultural Democracy,” where participation is conceptualized as far more active than the purchase of a ticket followed by passive viewing or experiencing of an event produced by professionals. At the Carnival, the floats are produced largely by community groups, assisted by technical specialists, in the months leading up to the event itself. In terms of the SACO framework, these values could fit into the “Human Capital and Professional Capacity Building” theme, which focuses on the skills and experiences gained by artistic producers and others involved with the supply-side of the event. However, the strong community focus of the Carnival, and its role in bringing diverse community groups together, could also be seen as part of the “Social Cohesion and Community Development” theme. Evaluation of the social goals of the event was conducted using qualitative techniques, taking into account the opinions of the intended beneficiaries themselves. After the Carnival, two focus group discussions with representatives of the community groups involved with producing the Carnival floats and displays were held. The focus group discussions explored two main areas, linked to the event aims: The Carnival’s role in showcasing cultural diversity and building social cohesion; and the development of professional capacity of creative entrepreneurs. Discussion included both ways in which the Carnival was succeeding and ways in which it could be improved. Discussions were transcribed and organized into “themes”—particular narratives organized around specific (sometimes implicit) topics that emerged from the data (Vaismoradi et al., 2016). The two major themes that emerged, given the guiding questions, were then divided into subthemes (Table 4). For example, within the cultural diversity and social cohesion theme, topics, or subthemes, that emerged were related to community inclusion, participation, recognition, and the importance of leadership and coordination. Within each of these subthemes, areas where the Snowball and Antrobus 15 community participants were of the opinion that the Carnival was meeting its goals (referred to as “positive dimensions”) as well as areas where things could be improved (referred to as “negative dimensions”) were expressed. Results showed that, for many participants, the feeling of being included and recognized, as well as the opportunity to participate, both formally in the event itself, and informally during preparations, were key contributors to their perceptions of the value of the event. Challenges identified in this area were often related to practical issues, such as ways of overcoming geographical problems caused by the fact that many of the communities who participate in the Carnival live in outlying areas and, coming from lower-income households, were not easily able to pay for transportation into the city center where the Carnival was held. Another key subtheme was recognition: Community participants greatly valued the sense of their worth and importance that participation in the Carnival engendered (positive dimension) but noted that holding the event in the city center made it inaccessible to lower-income communities (negative dimension). It was suggested that facilitating greater media coverage of the event (including social media) could play a role in increasing accessibility. The development of professional capacity among creative entrepreneurs, linked to developing sustainable careers in the creative economy, is also an important festival aim. Key subthemes here were the opportunities provided to develop through collaboration, both with those employed as specialists to help participants (such as choreographers and costume designers), but also between participants themselves. A challenge for many cultural events is that they take place over a short space of time, while career development and training are longer-term goals. Suggestions were made to extend the impact of the Carnival by, for example, developing and sharing a database of participants and their skills for the facilitation of future collaboration. Qualitative research, such as using focus group discussions with thematic analysis, may be extremely valuable in assessing the extent to which a cultural event is meeting its goals. Audience opinion surveys are most useful where “participation” is conceptualized as an audience that is “consuming” a cultural experience produced by professionals (cultural democratization paradigm). For the cultural democracy paradigm where participants are seen in both roles (sometimes called “prosumers”), more detailed, qualitative analysis may be more appropriate. Economic impact Tourism events attract “new” money into the impact region and create indirect and induced impacts through re-spending of the initial injection (the multiplier effect) that can be measured using economic impact studies. In terms of the SACO framework, economic impact fits into the “Inclusive Economic Growth” theme. Economic impact studies are prone to many forms of bias (see Snowball, 2008) but are popular because they can provide quantifiable evidence of the financial value of an event to the host city or region. However, economic impact results depend very much on the characteristics of the event itself (length, number of nonlocal visitors, sponsorship from outside the impact area), as well as the economic structure and reputation of the host city. For example, most researchers argue that expenditure by local residents should not be included in economic impact studies. The same applies to sponsorship from inside the impact area, since it is likely that, even if the event had not taken place, this money would still have been spent in the impact area, on something else (Crompton, 2006; Crompton et al., 2001; Snowball, 2008). A high proportion of local attendees at a festival is thus likely to result in a lower economic impact number. 16 Tourism Economics XX(X) Determining the number of attendees is one of the most important tasks in festival research, both for determining the social and the economic impact of the event. There are a number of methods available for calculating attendee numbers based on, for example, ticket sales and crowd density counts. For MACUFE, the ticket sales method was used, and it was estimated that the 10-day event attracted about 17,000 tourists, who stayed for an average of 4 days each (68,000 visitor days), as well as 8200 local residents. For the Cape Town Carnival, crowd density counts, survey data on the average number of hours that each person participated, along with organizer data were used. It was estimated that the event attracted an audience of 44,900 people. However, only 13% (5 800) were tourists, and only 41% of these (2400) were in the city specifically to attend the Carnival. The economic structure of the host city also affects the economic impact through the local capture rate (i.e. the percentage of visitor and organizer spending that is “captured” in the local market for at least one round), as well as the multiplier size. Cape Town, which has a larger population and more diversified economy, has a higher capture rate and spending multiplier than Bloemfontein/Mangaung. However, MACUFE attracts a much higher percentage of nonlocal visitors and takes place over a much longer time period than the Cape Town Carnival (Table 5). The South African Festivals Economic Impact Calculator (developed by Seaman and Snowball for the SACO) was used to calculate the economic impact of both events. The total economic impact of MACUFE 2016 was calculated as R103.4 million and R41.1 million2 for the Cape Town Carnival. The differences can be explained by both event characteristics (such as duration and the focus on tourists versus local residents)and the structure of host cities themselves. Another important factor to consider is the sources of event sponsorship. For example, MACUFE received a total of public sponsorship from national and provincial government of R44 million,3 with a much smaller proportion (R6m) coming from private sources. The Cape Town Carnival received R13 million4 in sponsorship, of which only R3.3m was from national or provincial public sources. In comparing the public “returns” from these two cultural festivals, one has to keep in mind not only their very different characteristics and aims but also the source of their funding. In the case of MACUFE, the public spending to economic impact ratio was 1:2.4, while for the Carnival it was 1:12.5. Despite the smaller size of the Carnival, it might be considered a better public investment in some ways, given how little the public sector had to invest per Rand of additional economic impact. In terms of maximizing the economic impact from cultural tourists, larger cities (with bigger multipliers and fewer outflows) can be more successful in attracting audiences and retaining tourist spending. However, bigger cities (like Cape Town), which have larger home populations and other cultural attractions, also tend to result in higher proportions of local event attendees and fewer people who are in the host city specifically to attend the event, both of which decreases tourist numbers and economic impact. Smaller cities, like Mangaung, may have more outflows and lower multipliers, but as long as they have the infrastructure to provide for tourists, and the event is wellmarketed enough to attract tourist numbers, may have a larger regional economic impact from cultural events because of a lower proportion of local attendees and a higher proportion of visitors to the city who are there specifically to attend the event. Conclusions While cultural tourism has become a popular part of local economic development strategies, the sustainability of the festivals that drive tourism depends on their careful management and monitoring, not only in terms of their ability to attract tourists but also in terms of the impacts on local populations. The rising number of festivals in South Africa, as in other countries, has led to greater Snowball and Antrobus 17 competition for tourists between various host cities and an increasing emphasis on valuation of the impacts of such events. In addition, funders in postapartheid South Africa, as in other places, are starting to demand that tourist events also consider their social impact, especially in building more diverse audiences, enhancing access and participation, and building social cohesion. This article demonstrated ways of applying a multidimensional valuation framework to two very different South African city festivals, based on the monitoring and evaluation framework developed by the SACO. Rather than focusing on economic impact alone, the research included examples of how to assess audience diversity, as well as the more difficult-to-measure social and intrinsic cultural values associated with cultural festivals. The article thus aimed to add to the theoretical literature on the measurement and valuation of cultural events that attract tourists, as well as suggesting practical ways in which such tools could be applied. By contrasting two cultural festivals in two cities, the article also contributes to the research on the characteristics of host cities and events that are likely to be more or less successful in attracting cultural tourists. The results suggest that longer festivals in medium-sized cities, as evidenced by the MACUFE music festival, can be successful in attracting cultural tourists and generating economic benefits for the region. MACUFE takes place over 10 days and is organized around traditional ticket shows. It has a large budget and receives much funding from public sources. It has been successful in attracting a large number of cultural tourists to the city, most of whom are from higher income and education groups, and are there primarily to attend the festival, which results in a large economic impact. While not diverse in racial terms, MACUFE audiences are culturally diverse when home language is considered. MACUFE helps to increase the supply of African music, which is associated with increased QoL for festival attendees. Shorter festivals in big cities, especially those with a strong focus on local communities, can be successful in attracting audiences and reaching social goals but may not be able to attract many cultural tourists and are thus likely to have a smaller economic impact. For example, the Cape Town Carnival is a 1-day event with a strong social and community focus, which is held in a large metropolitan area. It attracts a large audience, but most are local residents from lower income and education groups, not cultural tourists specifically in the city to attend the Carnival. It depends mostly of private funding and has much lower economic impact than MACUFE. However, it has important social and skills-development goals with regards to supporting local communities, and there is evidence that the beneficiaries regard the Carnival as at least partly successful in doing so. Moving forward, it is argued that flexible valuation frameworks that make use of a wider variety of indicators are likely to be more valuable to both event organizers and funders than singleindicator figures like economic impact. In managing and promoting cultural events, careful thought needs to be given to the balance between attracting cultural tourists and maximizing economic impact, versus the equally important social benefits to local communities. Authors’ note The original reports on which this research is based were produced for the South African Cultural Observatory, which is a national research organization funded by the Department of Sport, Arts and Culture, Republic of South Africa. Declaration of conflicting interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. 18 Tourism Economics XX(X) Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to acknowledge the support of the South African Cultural Observatory, which is funded by the national Department of Sports, Arts and Culture, South Africa. ORCID iD Jen D Snowball https://orcid.org/0000-0001-7497-6653 Supplemental material Supplemental material for this article is available online. Notes 1. It should be noted that some econometricians would argue that an average score is not a valid method of analysis, since the 1 to 5 scale is an ordered categorical variable not a continuous variable. However, since higher numbers (scores) do denote greater satisfaction, we would argue that averages do provide some useful information in this context. 2. R103.3m and R41.1 US$7.1m and US$2.8m, respectively, where 1USD ZAR14.5 3. US$3.03m 4. US$0.9m References Bohm S and Land C (2008) No measure for culture? Value in the new economy. Available at: http:// repository.essex.ac.uk/1261/1/no_measure_for_culture_pre_pub.pdf (accessed 18 October 2018). Bonet L (2011) “Cultural tourism”. In: Towse R (ed) A Handbook of Cultural Economics. 2nd ed. Cheltenham: Edward Elgar Publishing, pp. 166–171. Bonet L and Negrier E (2018) The participative turn in cultural policy: paradigms, models, contexts. Poetics 66: 64–73. Botha F and Snowball J (2015) Editorial: subjective well-being in Africa. African Review of Economics and Finance 7: 1–5. Cape Town Carnival (2016) Annual Review. Available at: capetowncarnival.com (accessed 15 September 2019). Crompton J (2006) Economic impact studies: Instruments for political shenanigans? Journal of Travel Research 45: 67–82. Crompton J, Lee S and Schuster T (2001) A guide for undertaking economic impact studies: the Springfest example. Journal of Travel Research 40: 79–87. Crossick G and Kaszynska P (2014) Under construction: towards a framework for cultural value. Cultural Trends 23: 120–131. Ebrahim S, Botha F and Snowball J (2013) Determinants of life satisfaction among race groups in South Africa. Development Southern Africa 30: 168–185. Ferguson A (2013) Symbolic capital and relationships of flow: Canada, Europe, and the international performing arts circuit. Theatre Research in Canada 34: 97–124. Galloway S (2006) Cultural participation and individual quality of life: a review of research findings. Applied Research in Quality of Life 1: 323–342. Glow H, Kershaw A and Reason M (2020) Leading or avoiding change: the problem of audience diversification for arts organisations. International Journal of Cultural Policy. DOI: 10.1080/10286632.2019. 1709060. Hand C (2018) Do the arts make you happy? A quantile regression approach. Journal of Cultural Economics 42: 271–286. Snowball and Antrobus 19 Herrero LC, Sanz JA and Devesa M (2011) Measuring the economic value and social viability of a cultural festival as a tourism prototype. Tourism Economics 17: 639–653. Holden J (2009) How we value arts and culture. Asia Pacific Journal of Arts and Cultural Management 6(2): 447–456. Holden J and Balta J (2012) The public value of culture: a literature review. European expert network on culture (EENC). Available at: http://www.eenc.info/wp-content/uploads/2012/11/JHolden-JBalta-publicvalue-literature-review-final.pdf (accessed 21 November 2017). Ivanovic M (2008) Cultural Tourism. Cape Town: Juta and Company. Jordan J (2013) The festivalisation of contemporary life. In: Newbold C and Jordan J (eds) Focus on World Festivals: Contemporary Case Studies and Perspectives. Oxford: Goodfellow Publishers, Chapter 2, pp. 6–18. Mangaung Metropolitan Municipality Integrated Development Report (2017) Available at: http://www.man gaung.co.za/wp-content/uploads/2017/04/DRAFT-Reviewed-IDP-2017-2022.pdf (accessed 14 June 2018). Michalos A and Kahlke P (2008) Impact of arts-related activities on the perceived quality of life. Social Indicators Research 89: 193–258. Mulcahy K (2017) Public Culture, Cultural Identity, Cultural Policy: Comparative Perspectives. New York: Palgrave Macmillan, Chapter 4, pp. 95–124. Nadotti L and Vannoni V (2019) Cultural and event tourism: an interpretive key for impact assessment. Eastern Journal of European Studies 10: 115–131. Quinn B (2010) Arts festivals, urban tourism and cultural policy. Journal of Policy Research in Tourism, Leisure & Events 2: 264–279. Quinn B (2016) Problematising ‘festival tourism’: arts festivals and sustainable development in Ireland. Journal of Sustainable Tourism 14: 288–306. Rogerson C (2008) Consolidating local economic development in post-apartheid South Africa. Urban Forum 19: 307–328. SACO (2016) A Framework for the Monitoring and Evaluation of Publically Funded Arts, Culture and Heritage. Report Produced for the South African Cultural Observatory (SACO). Available at: https:// www.southafricanculturalobservatory.org.za/ (accessed 15 October 2016). Snowball J (2008) Measuring the Value of Culture: Methods and Examples in Cultural Economics. Germany: Springer-Verlag. Snowball J (2011) Cultural value. In: Towse R (ed) A Handbook of Cultural Economics. 2nd ed. Cheltenham: Edward Elgar. Statistics South Africa (2015) Tourism. Report No. 03, 51-02-2015. Available at: http://www.statssa.gov.za/ publications/Report-03-51-02/Report-03-51-022015.pdf (accessed 29 March 2016). Statistics South Africa, Census (2011) Available at: http://www.statssa.gov.za/?page_id¼993& id¼mangaung-municipality (accessed 14 June 2016). Steiner L, Frey B and Hotz S (2015) European capitals of culture and life satisfaction. Urban Studies 52: 374–394. Throsby D (2010) The Economics of Cultural Policy. Cambridge: Cambridge University Press. Toerien D (2020) Tourism and poverty in rural South Africa: a revisit. South African Journal of Science 116: 72–80. University of the Free State Centre for Development Support (2015) MACUFE 2015: Socio-Economic Impact Assessment. Bloemfontein: CDS Research Report, LED and SMME Development. Vaismoradi M, Jones J, Turunen H, et al. (2016) Theme development in qualitative content analysis and thematic analysis. Journal of Nursing Education and Practice 6: 100–110. Van Heerden C and Saayman M (2018) Sustainability of a national arts festival: an application of a data envelopment analysis approach. Tourism Economics 24: 576–592. Van Hek M and Kraaykamp G (2013) Cultural consumption across countries: a multi-level analysis of social inequality in highbrow culture in Europe. Poetics 41: 323–341. 20 Tourism Economics XX(X) Werner B, Hayward J and Larouche C (2014) Measuring and understanding diversity is not so simple: how characteristics of personal identity can improve museum audience studies. Visitor Studies 17: 191–206. Willekens M and Lievens J (2016) Who participates and how much? Explaining non-attendance and the frequency of attending arts and heritage activities. Poetics 56(2016): 50–63. Wilson N, Gross J and Bull A (2017) Towards Cultural Democracy: Promoting Cultural Capabilities for Everyone. London: King’s Cultural Institute. Available at: https://www.kcl.ac.uk/Cultural/culturalen quiries/Towards-cultural-democracy/Towards-Cultural-Democracy-2017-KCL.pdf (accessed 19 January 2018). Author biographies Jen D Snowball is a professor of Economics at Rhodes University and has done some work on the cultural and creative industries for the South African Cultural Observatory. Her research in cultural economics has focused on the use of market and nonmarket valuation methods, especially as they apply to cultural festivals. Geoff G Antrobus is professor emeritus of Economics at Rhodes University and a researcher for the South African Cultural Observatory. His work in cultural economics has focused on the economic impact of festivals. His other areas of interest are agricultural and environmental economics.