**5. Discussion**

The present study was an empirical examination of temporal and seasonal distribution that is considered to be a possible solution for over-tourism. The possibility of distribution was tested by comparing willingness to pay between peak and non-peak times during night tourism activities at Suwon Hwaseong Fortress, a UNESCO World Heritage site. The contribution of this study is to expand the scope of the study of over-tourism to the overlooked field of night tourism.

The results and implications are as follows. First, in terms of frequency of night tourism visits, the group that also preferred the non-peak seasons group had higher visiting frequency compared to the group that only preferred peak seasons. There was, however, no difference between peak time preferred group and non-peak time preferred group. The result of the temporal comparison indicates that, regardless of preferred time zones, night tourism activities can only be offered until 10 p.m. That is, visitors to the study site are not familiar with various night zones. However, considering Mont-Shell Michel in France and the Taj Mahal in India, which are open from sunset to sunrise, it will be necessary to examine the possibility of expanding opening hours. The result of the seasonal comparison is in line

with substitutability that local residents and regional visitors can be made to visit during non-peak times (i.e., mornings, evenings, and weekdays) rather than at peak times [12].

Second, the comparison of willingness to pay for night tourism products between the group that preferred peak times and the group that preferred non-peak times showed that the willingness to pay was lower for all products except for tourism vehicles in the non-peak times group. The reason why willingness to pay of the non-peak times preferred group was high for tourism vehicles is that non-peak times are non-operating time zones at this site. Also, the number of seats and operating time zones of tourism vehicles are also fixed. Thus, there is an additional willingness to pay for currently non-operating time zones and to reduce inconveniences caused by overcrowding. There was also a positive willingness to pay for all attributes in the case of non-peak times preferred group. This suggests that visitors will still visit even if they are distributed to different time zones. It is thus necessary to expand gradually from the provision of services that do not cost a lot even during non-peak times. Tourism vehicles for which visitors indicate a high willingness to pay maybe a key product to attract tourists during non-peak times. This study expands the scope of night tourism with an expansion of temporal range, supporting previous studies which propose an expansion of night tourism targets [14,23].

Third, in relation to seasonal distribution, a comparison of the group that prefers only peak seasons and the group that also prefers non-peak seasons showed that the latter group showed a higher willingness to pay for all attributes. Specifically, willingness to pay was high for experience activities, tourism vehicles, and interpretation, suggesting a high preference for the avoidance of overcrowding inconveniences that occur during peak seasons for these contents. The high frequency of visits of the group that also prefers non-peak seasons seems to be influencing their high willingness to pay. These findings expand Hospers' [5] examination of the usefulness of temporal distribution by comparing visitors' willingness to pay, and provide practical implications for the applicability of temporal distribution.

### **6. Conclusions**

For sustainable night tourism, the management of over-tourism is important. This study examined the possibility of temporal and seasonal distribution as a solution for over-tourism in night tourism by estimating the willingness to pay for peak and non-peak times of visitors to a World Heritage site. This is important in terms of sustainable development of night tourism, and the implications of this study are as follows.

First, temporal and seasonal distribution were proposed as a solution for over-tourism in night tourism. Existing over-tourism research has been conducted in a case-by-case manner without empirical testing of methods that have been carried out under the general concepts of crowding or displacement, rather than considering site or visitor characteristics [2,5,9]. However, distribution criteria should be based on an understanding of both the characteristics of the destination and its tourism resources, as well as the target market. This study thus proposed a specific distribution method based on characteristics of night tourism. Night tourism refers to tourism at all other times during which day tourism occurs. With developments in transportation, one can easily leave destinations, and the benefits of night tourism such as overnight lodging do not always occur. Late night times (i.e., after 10 p.m.) or early morning times (i.e., earlier opening times) may be effective ways of temporal distribution. Seasonal distribution methods were also presented by comparing visitors who prefer non-peak seasons in addition to general peak seasons. These solutions can be effective in attracting additional tourists by providing differentiated tourism products that consider seasonal characteristics.

Second, the choice experiment method was used to examine in more detail the distribution of night tourism visitors. Choice experiments allow estimations of the MWTP according to each attribute and level of night tourism products. The findings suggest that there are restrictions on service provision for temporal and seasonal distribution, and that emphasis should be placed on interpretation, tourism vehicles, and experience activities during which the inconveniences of overcrowding are especially felt.

Third, the applicability of temporal and seasonal distribution was tested through an examination of visitors' willingness to pay during peak or non-peak times. The distribution of visitors to non-peak times is aimed at sustainable development of target resources [9] in terms of preventing peak season crowding, improving satisfaction, and reducing fluctuations in demand, through an easing of over-tourism in night tourism. In addition, it is possible to broaden the field of night tourism by targeting time zones during which night tourism content is not yet provided. Thus, in order to avoid crowding during peak times and lower quality tourism experiences, there is a need to provide both pleasant and differentiated experience activities during non-peak times for visitors.

Finally, similar to previous studies that examined the characteristics of potentially distributable groups [11,12,27], this study examined characteristics of the group of visitors that preferred non-peak times. According to Manning [12], visit frequency may still be high for local residents and regional visitors even if time zones are changed. Providing support for this, visitors who also preferred non-peak seasons had a high number of night tourism experiences at the study site (22.4 average visits vs. 13.2 average visits for peak times). The non-peak preference group can be seen as loyal customers for whom various strategies such as discounts, and time zone or season specific services and product development are needed. The results of this study indicated a high willingness to pay for night tourism experience activities, tourism vehicles, and interpretation, allowing specific distribution strategies to be established.

Given the exploratory nature of this study in which just the possibility of distribution was tested, the intention to visit based on specific services/products was not directly examined. For actualization of distribution, more detailed research is needed on so-called loyal customers who are likely to visit frequently. Through big data analysis, visit patterns by times/seasons should be obtained, and basic distribution data should be prepared in order to break down time zones and seasons in more detail. Since this study is about the specific case on over-tourism of night tourism, to generalize the results more researches are needed in other cases. In addition, to introduce a temporal distribution as a solution for over-tourism, the incentives for workers and local residents, who provide the services, should be considered. Despite theoretical and practical limitations, the present study is meaningful in that it pursues sustainable development through elimination of over-tourism in night tourism at a World Heritage site by testing the possibility of distribution in more detail using visitors' willingness to pay.

**Author Contributions:** Project administration and funding acquisition, H.S.; Literature Review, H.S. and M.K.; Conceptualization, Modeling and Survey Review, all authors; Survey and Data Collection, H.S. Methodology and Data Analysis, C.P.; Writing—original draft preparation and editing, all authors. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by 2019 Policy Research of Suwon Research Institute, SRI-Policy-2019-21.

**Acknowledgments:** We thank the anonymous reviewers for their insightful comments and suggestions to enhance the quality of the manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.
