Evaluating the efficiency of Korean festival tourism and its determinants on efficiency change: Parametric and non-parametric approaches

Abstract

Local festivals may leverage local specialties and various historical, cultural, and artistic resources throughout their respective regions to attract tourists, inducing positive economic impacts. In this study, this paper is a first attempt to analyze the relative efficiency of local festival tourism by using parametric and non-parametric approaches with the data from local festivals held in Korea from 2015 to 2018. We also deal with the efficiency determinants of each typology of festivals by employing a truncated regression with double bootstrapping. Results showed that the leading sources of inefficiency were primarily embedded in pure technology inefficiency, while the principal operational drivers posed different effects depending on the typology of festivals. These insights have important practical implications for the local festival organizing committees and operators in Korea and are helpful in developing tailored operational strategies to maximize the efficiency among different typologies of festivals.

Introduction

Festival tourism, which refers to the attraction of visitors to local festivals by using a local region's tourism resources, is one of the fastest-growing fields in the tourism industry (O'Sullivan & Jackson, 2002; Getz & Page, 2016). Local festivals may leverage a variety of local specialties and the various historical, cultural, and artistic resources available in the region to attract tourists, inducing positive economic ripple effects, which can revitalize local economies, and creating new jobs (Getz, 2008; Lee et al., 2011; Mules & Dwyer, 2006, pp. 206–223). In particular, the representative local festivals have gained worldwide reputation such as Oktoberfest, Edinburgh International Festival, Sapporo Snow Festival, Rio Carnival, Burning-man festival, Gilroy Garlic Festival, Menton Lemon Festival, and Arena di Verona Opera Festival. These renowned events not only play a role in creating favorable national images but also draw millions of tourists from around the world, thereby creating enormous economic and social value to the individual nation (Alves, Cerro, & Martins, 2010; Zhang, Fong, & Li, 2019). For instance, Oktoberfest, which has become known as the world's largest beer event, drew approximately 7.2 million tourists to Munich in 2019, where they consumed 6.9 million liters of beer, 550,000 chickens, and 172 cows, thus amounting to an economic value of 1.2 billion euros (www.oktoberfest.net). In 2014, Japan's Sapporo Snow Festival generated direct and indirect economic effects estimated at 41.9 billion yen with 2.4 million tourists, which was 150 times greater than the 29 million yen investment made by the local government in order to host the festival (en.prothomalo.com/opinion/Sapporo's-snow-for-economic-benefits).

With the growing awareness that local festivals can improve regional images while boosting the local economy and further inducing the employment of local residents, festival events utilizing various regional arts and cultural resources are competitively held from around the world. As of 2019, there were approximately 1110 music festivals, 1093 film festivals, and 1930 festivals related to food and beverages in the United States alone (www.festforums.com). The numbers are even greater in China, where more than 5000 local festivals are held each year (Lu et al., 2009).

However, the recent surge in excessive competition among local festivals has been posing a serious threat to the overall sustainability of festival tourism (O'Sullivan & Jackson, 2002; Van Heerden & Saayman, 2018). According to Van Heerden and Saayman (2018), more than 600 festivals are held every year in South Africa, but despite the continually increasing number of events, the number of attendees has been plummeting. O'Sullivan and Jackson (2002) also empirically investigated that economic profits of the festivals decrease in proportion to the number of festivals.

In Korea, 884 local festivals were celebrated in 2019 (kto.visitkorea.or.kr), but a number of festivals, except a few successful local events, trigger relatively low economic effects compared to the investment made by their respective local government. This was mainly due to a lack of festival-specific visions or values and the poor operation management. Additionally, several local festivals resulted in low tourist satisfaction with poor festival programs, insufficient infrastructure, and service policies that did not meet the needs of tourists effectively. Accordingly, some controversies have recently been raised over whether or not to continue holding the local Korean festivals.

In this backdrop, study 1 employed parametric stochastic frontier analysis (SFA) to estimate the efficiency in four typologies of local festivals held in Korea from 2015 to 2018. Within the context of thin and short panel data, a simple and naive approach is to pool the data across different periods (Du et al., 2018). Accordingly, this study considered the period from 2015 to 2018 as a cross-sectional analysis, to pool the festivals of each typology for the period comparing them as a cross-section analysis and then compared the average festival groups. Local festivals can generally be categorized based on their materials, and the motivation of festival attendees; different festivals tend to set different goals and operational strategies, depending on the type of festival (Crompton & McKay, 1997; Getz & Page, 2016; Lee et al., 2004; Ma & Lew, 2012; Nicholson & Pearce, 2001; Scott, 1995). Previous studies investigated the different motivations of visitors attending various festivals with different characteristics. Scott (1995) compares visitors' motivation for visiting three festivals: the holiday lights festival, the bug fest, and the maple sugaring festival. They categorized motivational factors into six dimensions: event excitement, nature appreciation, curiosity, sociability, family togetherness, and escape from routine. Nicholson and Pearce (2001) highlight the diversity in motivation to attend different events using a comparative analysis of the visitors’ motivations to events such as food and beverage festivals, a country and music festival, and an air show. Ma and Lew (2012) argue that different festival types have different challenges. They classified Chinese festivals into four categories—local heritage, local modern, national heritage, and global modern—according to historical and geographic dimensions of the events. Local identity, uniqueness, fun/liminality, and authenticity were mentioned as important concerns that affected the dimensions. They further state that for event managers to plan the festival successfully, it is necessary to focus on these issues, which are related to the different types of festival. In Korea, regional festivals are classified into four different typologies based on the main themes and contents of the festival, including culture and art (e.g., music, dance, opera, theater), local specialty products (e.g., beer, lemon, garlic, rice wine, tulip), ecological and/or nature-related (e.g., snow, purple butterflies, cherry blossom, whale), and historical and/or cultural heritage (e.g., shamanism, Confucianism). In study 1, we estimated the festival performance of four different typologies, based on the translog output distance function, and compared the results using two approaches: parametric SFA and non-parametric DEA (data envelopment analysis).

In study 2, we measured metafrontier technical efficiency (TE*), technical efficiency (TE), and technology gap ratio (TGR) by using metafrontier DEA methodology to compare the efficiencies among different festival groups with heterogeneous production technologies. We also analyzed the causes of inefficiency, returns to scale, and fluctuation patterns of metafrontier index values, thereby providing subsequent insights for festival organizers and operators. Furthermore, we demonstrated the efficiency determinants of each typology of festivals by employing truncated regression with double bootstrapping, as suggested by Simar and Wilson (2000, 2007a), for identifying the key drivers of efficiency variation. Accordingly, we attempted to figure out major operational variables that affected efficiency change in each typology of local festivals, while addressing strategic operational initiatives tailored to individual festival groups in order to improve contemporaneous efficiency. The research issues of interest in this study are:

• Are there any differences in technical efficiency estimates using parametric and non-parametric approaches among the different typologies of local festivals in Korea?

• What are the major sources of inefficiency observed among different local festivals group?

• What are the fluctuation patterns of TE*, TE, and TGR among different local festival groups?

• What are the festival-specific determinants that affect technical efficiency of each festival group?

Through studies 1 and 2, we thus offer three theoretical and managerial contributions:

• While the vast amount of previous studies have measured the efficiency for specific aspects of tourism, such as restaurants (Alberca & Parte, 2018; Reynolds & Thompson, 2007), airlines (Choi, 2017; Merkert & Hensher, 2011; Yu et al., 2019) and hotels (Assaf et al., 2010; Barros, 2005; Lado-Sestayo & Fernández-Castro, 2019; Oukil et al., 2016), previous applications of SFA and DEA models in the field of festival tourism have been scarce. In particular, this paper is a first attempt to analyze the efficiency of festival tourism via parametric and non-parametric approaches.

• According to the results of the metafrontier analysis, there were significant mean differences in metafrontier technical efficiency among festival groups. The average TE* in the local specialty product festivals group was the highest, while the TE* of the ecological and/or natural-related group was the lowest. Furthermore, the main driver of inefficiency in local Korean festivals was mostly stated to be pure technology inefficiency, which was further negatively affected by operational inefficiency. In particular, most ecological and/or natural-related festivals showed pure technology inefficiency, having the lowest TE* values. Thus, this group requires innovative operating policies and strategic benchmarking initiatives to improve its operational efficiency.

• Lots of festivals in the local specialty product festivals group have relatively longer histories and offer well-known local food traditions that imprint stronger images upon potential tourists. This widespread recognition produces a more stable festival operation, compared with those in other groups, and ultimately leads to more consistent metafrontier index values.

• This study empirically identified which operational factors affected the efficiencies for four typologies of the local festival. Our results showed the efficiencies of the categorized festival groups are influenced by different explanatory factors, thus suggesting that festival organizing committees and operators require tailored operational strategies to maximize their efficiency.

The rest of this research is structured as follows. The previous literature on festivals is presented in section 2. The research model and empirical data used in this study are presented in Section 3. Section 4 estimates the parametric efficiency for four local festival parties. In section 5, we performed the metafrontier analysis and compared results to those of SFA. Using the Simar and Wilson approach, we evaluated the impact of contextual variables on efficiency for the festival group in section 6. Section 7 discusses the theoretical and practical implications as well as study limitations.