The tourism and economic outcomes of hosting a sporting mega-event: a case study of the Singaporean F1 Grand Prix

Research Paper Academic year 2023-2024 The tourism and economic outcomes of hosting a sporting mega-event A case study of the Singaporean F1 Grand Prix Alex Carreras Lorenzo Under the supervision of Professor Michael IMPINK Major in Economics & Finance 2 3 Abstract The effects of hosting sporting mega-events have been extensively studied, particularly with a focus on prominent events such as the Olympic Games or the FIFA World Cup. However, less attention has been given to the impact of Formula One events on the tourism indicators of the host regions. This paper reviews the existing literature and employs multiple regression models to analyse the influence of the Singaporean Formula One Grand Prix on tourism figures. The results from the models indicate that the event has not significantly affected tourism arrivals and receipts in the Asian city-state. Through this analysis, the study aims to provide insights for future policy and decision-making for cities considering hosting similar events, such as the upcoming Madrid Formula One event. 4 5 Table of Contents 1. Introduction ............................................................................................................. 6 1.1 Public investment: a catalyst for economic development ..................................... 6 1.2 Research Question and objectives ...................................................................... 7 2. Literature Review .................................................................................................... 9 2.1 Impact of hosting a world non-sporting event ...................................................... 9 2.2 Impact of hosting a world sporting event............................................................ 10 2.2.1 Main benefits of organizing sporting mega-events .................................................... 10 2.2.2 Main disadvantages of organizing sporting mega-events .......................................... 11 2.3 Impact on tourism of hosting 3 concrete sporting events ................................... 12 2.3.1 Effect of hosting the Olympic Games ......................................................................... 12 2.3.2 Effect of hosting the FIFA World Cup ......................................................................... 14 2.3.3 Effect of hosting an F1 Grand Prix ............................................................................. 16 3. Hypothesis Formulation ....................................................................................... 19 4. Case study: Singaporean F1 Grand Prix ............................................................. 21 4.1 Overview ........................................................................................................... 21 4.2 Research Design ............................................................................................... 22 4.3 Data summary ................................................................................................... 25 4.3.1 Summary of the number of visitor arrivals .................................................................. 26 4.3.2 Summary of the amount of tourism receipts .............................................................. 30 4.4 Results .............................................................................................................. 33 4.5 Interpretation of results ...................................................................................... 40 5. Conclusions .......................................................................................................... 41 6. Bibliography .......................................................................................................... 43 6 1. Introduction 1.1 Public investment: a catalyst for economic development It is generally accepted that the world is nowadays living the most dynamic, fastgoing and intense era of globalization ever witnessed. In this context, cities and nations have to pursuit urban competitiveness to keep the pace of the environment changes. To achieve it, they take advantage of public investment. It represents a cornerstone of economic strategy, enhancing competitiveness and improving the overall well-being of citizens. Public investment refers to the allocation of public funds into initiatives and projects that are expected to yield benefits in the long-term. These investments are part of the government spending, and they are important to achieve a sustainable economic development. Ultimately, it plays a key role to improve the quality of life of citizens worldwide by offering more public services like transportation, safety or health care. For example, public investment encompasses expenditure on physical infrastructure like bridges, roads or public transportation system, as well as investment in human capital through healthcare and education. Large public investments are crucial in setting the stage for sustainable development by ensuring that the foundational assets, such as energy infrastructure, education system or affordable housing can support future growth. An example of how a public investment sets the stage for future growth is the Sapa airport project, located in the northern region of Vietnam. With a financial commitment that exceeds 300 million dollars, it will enhance the connectivity of the region, meeting the increasing demand for air travel and stimulating economic development by boosting regional accessibility. Public investments have the potential to stimulate economic activity, increase general efficiency and attract private investment. The OECD countries spent an average of 3.3% of their GDP in public investment in 2020, showing its importance worldwide. For instance, the European Investment Bank funded projects across Europe for a value of €60 billion in 2020, mainly focusing on 7 climate action, infrastructure and innovation. Investing in sports is an important part of the public investment. Even if there is limited data on the proportion of the public expenditure allocated to sports, most of the countries have systematically bet for sporting events and for improving their infrastructure, such as the US government spending $700 million for the construction of the Atlanta’s football new stadium. A highly visible form of public investment is the decision to host a mega-event. Roche (2000) describes these events as significant large-scale occurrences that are widely popular and that have international importance. They can include cultural, sporting and commercial aspects. Sporting events attract massive global audiences, providing a unique opportunity for cities and countries to showcase themselves on the world stage. These events help enhance their international image, offer new opportunities for local communities, and bolster their diplomatic standing. Cities vie to host these major events, such as the Olympics or Formula One Grand Prixes, hoping for economic gains through increased tourism and urban renewal. Such events yield both direct economic benefits, like ticket sales and sponsorships, and indirect benefits, including growth in hospitality and retail sectors while leaving a legacy of urban transformation. The discussion on the economic repercussions of hosting mega-sports events has captured the attention of academics, policymakers and the general public. Proponents argue these events bring substantial economic advantages, while critics point to examples where the outcomes did not meet the expectations, burdening host cities with debt and underutilized infrastructure. 1.2 Research Question and objectives In this context, it is relevant to further explore if organizing an international sporting event is economically beneficial for the hosting region. That is, to check if the celebrations when the region is announced to host the event are justified or if, on the contrary, hosting sporting events has some disadvantages that can worsen the economic situation of the country. 8 Therefore, the specific research question of the paper is: How does hosting a sporting event influence economic and tourism growth in the host country? In order to assess this study, we will start with a literature review section where we will examine the relationship between hosting major events and the economic impact for the hosting region. We will first focus on the main advantages and drawbacks of organizing large events. Then, we will deep dive in three types of sporting events: the Olympic Games, the FIFA World Cup and, finally, Formula One events. Following the literature review, we will state some hypothesis, and we will try to confirm or reject them through a concrete case study: the Singaporean Formula One Grand Prix. This case study will complement the existing literature by filling a gap: most of the studies about Formula One focus on European regions and none of them emphasises on the Asian city-state data. 9 2 Literature Review 2.1 Impact of hosting a world non-sporting event Research on the economic impacts of hosting non-sporting mega-events, such as World Expos, cultural festivals, and international conferences, reveals a range of outcomes. Some studies show that these events can bring significant economic benefits, including increased investment, job creation, and a heightened international profile. For example, the Bureau International des Expositions (BIE, 2010) reports that World Expos often lead to considerable infrastructure development and urban renewal in host cities, with long-lasting economic advantages. On the other hand, some research suggests that these anticipated economic benefits do not always come to fruition. Jones (2001) points out that while the Edinburgh International Festival boosts local economies through tourism and cultural spending, the high costs associated with organizing such events can sometimes outweigh the financial returns if not managed efficiently. When it comes to tourism, non-sporting mega-events are often credited with attracting large numbers of visitors, thereby boosting tourism receipts and raising the international profile of the host city. Richards and Wilson (2004) found that events like the European Capital of Culture significantly increase tourist arrivals and spending in host cities, leading to both immediate and longer-term tourism growth. However, these positive impacts are not guaranteed. A study by Getz (2010) points out that the tourism benefits of cultural events can be overestimated if the event fails to attract a significant number of international visitors or if the increase in tourists leads to overcrowding and strains on local infrastructure, potentially deterring future visitors. 16 For the 2014 FIFA World Cup in Brazil; Baumann and Matheson (2018) conclude that it attracted one million visitors. They claim it was a substantial result since it occurred during the typically low-tourism months in Brazil. Moreover, they present in interesting point that differentiates the World Cup from the Olympics. They argue that on-field results have a huge impact on tourism attraction during the event. Almost a quarter of the increase in tourism was due to the fact that their neighbour country Argentina reached the final game. Although the tourism arrivals grew by 50% compared to previous years during the months of the event, it was only a short-term effect. Meurer and Lins, 2017, showed that the post-World Cup month numbers are similar to the previous years. 2.3.3 Effect of hosting an F1 Grand Prix The economic impact of hosting Formula One Grand Prixes has garnered limited attention in the literature, so the literature review focuses more broadly on more general large sporting events. Proponents argue that hosting the race can bring substantial revenue through increased tourism, sponsorship deals, and global media exposure, driving local economic activity. The influx of international visitors often leads to heightened spending in the hospitality, retail, and service sectors. Additionally, hosting the Grand Prix frequently involves infrastructure upgrades that can enhance a region's long-term connectivity and appeal. However, critics highlight the high costs and financial risks of organizing these events, noting that short-term gains may not always justify the public investments made. Formula One is a globally followed sport, with more than 420 million people watching it every year (Jenkins et al. 2016). These amounts are comparable to the mega-event already mentioned: the Olympic Games and the FIFA World Cup. As a difference from the previous events, F1 Grand Prixes are held every year in the same location. This facilitates the usage of infrastructure investments, whose costs are significantly lower than the Olympics or World Cup ones (Alm, Solberg, Storm and Jakobsen, 2014). This phenomenon, combined with the worldwide attention, can serve to boost tourism and increase economic activity in the host regions (Remenyik and Molnar 2017). Consequently, it is interesting to examine these effects. 17 Just as a quick introduction to the sport, Formula One was acquired by Liberty Media in 2016 for $8 billion and had a turnover of $1.8 billion in the following year. This year (2024), the championship comprises 24 races distributed in 20 countries (there are 3 races in the United States and 2 in Italy). From Australia to Abu Dhabi and from Canada to Brazil, Formula One is a truly global sport. The effect of Formula One goes beyond the emotion of motorsport. Countries and nations have realized it can be a powerful advertising campaign, so it comes as no surprise that, apart from the British Grand Prix, all current F1 races receive considerable public funding (Jenkins et al, 2016). A common methodology to study the economic effects of F1 are input-output modelling (Jasmand and Maenning 2008). An example of this method can be found regarding the Chinese Grand Prix. They estimated that the output was about $30.6 million, the income was approximately $11.2 million and that it created 1,400 full-time jobs. Even if foreign spectators represented only 6% of the total, they accounted for more than 25% of the total expenditure. Consequently, organizers should focus on attracting international attendees, put in other words, attracting more tourists. Another example concerns the Australian Grand Prix held in Melbourne. A report from the consultancy firm Ernst & Young (2011) stated that the 2011 GP increased the Victorian gross state product by approximately $35 million, creating about 350-400 new jobs. These studies provide an optimistic view regarding the effect of hosting a Formula One Grand Prix. However, critics state that they leave out the cost of the event (Taks et al. 2011). Besides, the methodology used in these analyses has also been debated. Some authors have mentioned they use exaggerated multipliers (Matheson 2009, Siegfried and Zimbalist 2000) or that most of them do not consider substitution and crowding out effects. Apart from that, opportunity costs and environmental impacts are not usually taken into account (Cairns 2008). Bearing this in mind, some authors have used cost-benefit analyses (CBA) to estimate the effect of an event by taking all benefits and costs into account. Campbell (2013) applied CBA to the Australian F1 Grand Prix and concluded it 18 triggers losses of about $50-60 million per year. Therefore, they recommend the Victoria State to discontinue the event. Storm, Jakobsen and Nielsen (2019) use panel regression data to analyse the economic effects of Formula One races in European regions. They cannot prove that it brings positive consequences on employment, per capita GDP or tourism. Conversely, adverse outcomes often appear 3 to 4 years following the event. One of the causes for this lagged and negative impact can be the misallocation of public resources. Hosting a Formula One event needs substantial financial commitments from the host city or country in the form of subsidies or infrastructure investment (track construction, accesses, etc). This indicates significant opportunities costs, because public funding could be allocated in a more profitable initiative. 19 3. Hypothesis Formulation After reviewing the existing literature, we can synthetize some conclusions. This will allow us to formulate a series of hypothesis, that will be confirmed or rejected through the upcoming analyses. There is a general consensus about the fact that mega-events tend to present lower economic impacts than what is initially expected. This discrepancy can be attributed to an inclination to exaggerate the anticipated benefits to impress the committees responsible for selecting the host city. Another contributing factor could be the use of flawed or incomplete methodologies. Regarding the Olympic Games, most academics agree that they are not economically profitable in most of the cases. It is only in concrete editions that the Games turn out to be economically profitable. The huge investment needed, the complexity of using the newly-created infrastructure and the “one-shot” nature of the event can be reasons to explain this phenomenon. However, they can be useful when it comes to tourism influence. Not only can they serve to “put a city in the map” as it was the Barcelona 1992 case, but it can also boost the shortterm tourism. So, all in all, although they can present some tourism benefits specially in the short-term, the literature agrees that the Olympics and the World Cup are not massive profitable events. Opportunity cost of public funds, crowding out and substitution effects are more reasons than demonstrate this fact. Shifting focus to the specific case of a Formula One Grand Prix, the impact on tourism from hosting the race appears negligible. Few analyses have considered some European and Australian Grand Prix events. Formula One has become more global recently, with races held all around the world. These newer events have not been examined in detail. Based on the literature, I present two hypothesis that try to fill a gap in the existing papers. Through some regression analysis that will be detailed in the next section, this paper will try to confirm or reject the following hypothesis related to tourism impact of hosting sporting mega-events: 20 • Hosting a Formula One event significantly impacts the number of tourism arrivals during the month the event takes place compared to a typical month • Hosting a Formula One event significantly impacts the tourism receipts that the host region generates during the month of the event compared to a typical month This paper aims to enrich the existing literature by studying the case of the Singaporean Formula One Grand Prix effect on the city-state tourism. It is interesting to check if Singapore most important motorsport event’s impact is aligned with other Formula One races or if it is more similar to other mega-events like the Olympics or the FIFA World Cup. 21 4. Case study: Singaporean F1 Grand Prix 4.1 Overview In order to test the hypothesis abovementioned, this paper examines a particular case study: the Singaporean Formula One Grand Prix. This Grand Prix was held for the first time in 2008 and has been a fixed event in the motorsport calendar until the present days. It has been held once a year, usually in September, and was only cancelled during 2020 and 2021 as a consequence of the COVID-19 pandemic. The main characteristics of this event, and what really distinguishes it from other Grand Prixes, is the fact that it is based on a urban track around the streets of Singapore. Besides, it is the first night race that was established in the Formula One calendar. There are different reasons for selecting this Grand Prix. On the one hand, it has not been widely explored by the existing literature. On the other hand, the Singaporean Grand Prix can serve as an interesting reference for upcoming events like the one in Madrid. Figure 1: Layout of the Singaporean Formula One Grand Prix 22 Formula One events has been less examined than other mega-events such as the Olympic Games or the FIFA World Cup. Apart from that, most of the articles and papers covering the effect of hosting Formula One events on economic and touristic aspects put the focus on European events or on the Australian Grand Prix. Therefore, by analyzing this event we contribute to the expansion and completeness of existing papers regarding the correlation between mega-events and economic and touristic impact. Moving onto the second argument, the local authorities of Madrid, together with F1 management, have recently announced an agreement to host an urban Grand Prix in the Spanish capital starting in 2026. Madrid is pursuing a global strategy to boost the brand and image of the city at a global level. Singapore is a great example in this sense, as it has been following this objective for the last decades. Therefore, analyzing the impact of the F1 in Singapore will give an idea of the possible consequences for Madrid and any potential new tracks that can join the already packed Formula One calendar. Having lived in Singapore for half a year, I realized the importance of the event in the reputation of the city and on the national pride feeling that Singaporeans have. 4.2 Research Design To confirm or reject the hypothesis derived from the literature review, we will analyze two dependent variables that represent the tourism impact of hosting a sporting mega-event. These variables are the monthly number of arrivals to Singapore (through air, land and sea) and the monthly tourism receipts, which can be defined as the amount of money that tourists spend during their stay in the city-state of Singapore. The time horizon analyzed goes from January 2007 to December 2023. By analyzing these variables, not only do we consider the quantity of people that visit the city, but we also take into account the amount of money they spend. This may seem a bit obvious but it is actually quite essential. The number of visitors is important because it gives a sense of the magnitude of the tourism, but more people does not necessarily mean more income. 23 The statistical methodology applied is based on multiple linear regressions. This will allow us to isolate the effect of different variables and ultimately observe the marginal impact of hosting the Formula One Grand Prix on the dependent variables. To do so, we will need to control for some variables. For this reason, there will be different models for each dependent variable (tourism arrivals and tourism receipts). The first model is the pooled OLS. It only regresses the independent variable against the dependent one. Our independent variable is a binary one that takes value 1 if the Singaporean F1 event took place during that month and 0 otherwise. The next two models will include economic control variables (mainly GDP and unemployment rate). Then we will have two more models that include time fixed effects. Firstly, we will control for year fixed effects. This is important to isolate circumstances that occurred during a year and that may cause a significant change in the output. A clear example of that would be the financial crises in 2008-2010. If we did not control for these variables, the coefficients would possibly be inaccurate because of the omitted variable biased effect. Finally, we will add month fixed effects to control for the time trend. These new control variables will be useful to isolate the impact of circumstances that change the output just because of the month in which they occur. An intuitive example would be the fact that during the high-season (July and August), many people have some holidays and it is therefore reasonable to see more travels and more tourism arrivals. By including the month-level fixed effect we isolate this phenomenon. The most complete models are certainly the ones that include both economic and time fixed effects, which can be described through the following equations: 𝑌𝑖= 𝛽0+𝛽1·𝐹1_𝐸𝑣𝑒𝑛𝑡𝑖+ 𝛽2·𝐿𝑜𝑔_𝐺𝐷𝑃𝑖+𝛽3·𝐿𝑜𝑔_𝑈𝑛𝑒𝑚𝑝𝑙𝑜𝑦𝑚𝑒𝑛𝑡_𝑟𝑎𝑡𝑒𝑖 +∏(𝛽𝑗·𝑌𝑒𝑎𝑟𝑗+2) + ∏(𝛽𝑘·𝑀𝑜𝑛𝑡ℎ𝑘) Where Log denotes (natural) logarithm, i represents each month from January 2007 to December 2023, j is an index to represent years from 2008 to 2023 and k an index to represent months from 02 to 12. ∏ is the symbol for the product 24 operator. For a proper construction of the fixed effects variables, one month and one year is left aside as they serve as the baseline. Each β is the coefficient that relates the independent variable with the dependent variable. The dependent variables Yi are Log_Visitor_Arrivalsi and Log_Tourism_Receiptsi. The independent variable is F1_Eventi. For the control variables: • GDP is a continuous logarithmic variable that accounts for the monthly GDP of Singapore • Unemployment rate is a continuous logarithmic variable that accounts for the monthly percentage of unemployment in Singapore • Yearj+2 are binary variables that equal one if the datapoint belongs to that specific year • Monthk are binary variables that equal one if the datapoint belongs to that specific month The model tries to determine the relationship between hosting the Singaporean Grand Prix and the impact on tourism. However, there are some potential limitations that should be taken into account when it comes to draw robust conclusions. Even if we find a correlation between the dependent and independent variable, it does not necessarily mean there is a causation effect. Maybe the correlation is due to some other unexplored factors that are not captured in the model. Apart from that, the study may have less “external validity” since it is related to a single specific event. Therefore, the results are limited to Formula One Grand Prixes and similar events like other motorsport competitions such as MotoGP. This analysis aims to contribute to the existing literature review, and it should be considered together with other events to draw a solid relationship. Thinking about potential further works, not only can more events be studied, but also more dependent variables. Here we examine two variables as a representation of tourism impact, but more variables could be included to strengthen the results. If the goal is to extract conclusions for similar upcoming events, the organizers should also consider the cultural, geographical and economic differences between Singapore and the new location, as they may differ in a sense that significantly changes the outcome. 25 4.3 Data summary The study uses data from the official sites of Singapore, which include the Singapore Tourism Board, the Changi Airport Group and the Singapore Ministry of Manpower. All the analysis are conducted using the statistical software R. The geographical scope of the analysis is the city state of Singapore. As we have mentioned in previous sections, there are two dependent variables: number of visitor arrivals and the amount of tourism receipts. We use monthly data, which means there is one single datapoint for the whole city at the month level. The data ranges from January 2007 to December 2023 for the visitor arrivals and from January 2008 to December 2023 in the case of tourism receipts. This difference is mainly due to reliable data availability. So there are 204 datapoints for the first dependent variable and 192 datapoints for the second one. In order to get a first idea of the data that we are treating, here there is a summary of the different variables: Table 1: Summary statistics of the data 32 All years follow a similar evolution, showing an increase from January to October and then slowing down until the end of the year. This evolution seems to be logical as the last trimester of the year is part of the low-season period in Singapore. Because of COVID-19, 2020 and 2021 are below the rest of the years and 2022 shows some strong recovery from September onwards, as part of the recovery process once the pandemic crisis was solved. Again, we can look at the STL decomposition to better understand the behavior of the tourism receipts. Once we isolate the seasonal component, there is not a strong trend in the data from 2008 to 2020. Of course the pandemic shock is a massive change in the tendency but pre-pandemic levels are reached by the end of 2022. Figure 12. Seasonality in tourism receipts (yearly view) 33 4.4 Results After the first overview to the data, the next step is to run the multiple linear regression analysis. This will allow us to examine the relationship between hosting the Formula One event and the tourism indicators. As introduced before, there will be different regressions for each dependent variable. The differences between the models are the introduction of control variables. The next two tables show the results of the analysis. The first one refers to the number of visitors as the dependent variable and the second one refers to the amount of tourism receipts. In Model (1), we simply regress the dependent variable with the independent variable F1_Event. It is the binary variable that takes value 1 if the event took place in that month. Therefore, the coefficient of this raw is the most important to our study. It represents the impact of organizing the Formula One event. Nonetheless, including only this variable would be imprecise and uncomplete. We need to include control variables to isolate the effect of the event from other potential causes that impact the tourism figures. For this reason, in Model (2) we add the economic control variable of the GDP (in logarithmic form). Model (3) also includes the unemployment rate control variable. Figure 13: STL decomposition of the tourism receipts 34 To achieve more reliable and robust results, we also need to control for time trends, as we have seen in the data description that the data presents some seasonality. Model (4) includes variables that control for year fixed effects. Finally, Model (5) adds variables to control for month fixed effects. Both tables show the coefficient (β) associated with each variable and, in brackets, the corresponding p-value. 35 Table 2: Coefficient results for monthly number of visitor arrivals in Singapore 36 Table 3: Coefficient results for monthly amount of tourism receipts in Singapore 37 We also plot the diagnostic graphs, which allow us to ensure that the fundamental assumptions of the method (homoskedasticity and normality of the error term), are respected. This is essential to make sure that the OLS (ordinary least squares) methodology used to determine the Beta coefficients is accurate. The flat line in both residual plots confirms there is homoskedasticity and the diagonal shape in the Q-Q Residuals plot confirms that the errors follow a normal distribution. Therefore, the assumptions of the OLS methodology are respected. Figure 14: Diagnostic plots for the model with year and month fixed effects Figure 15: Diagnostic plots for the tourism receipts model with year and month fixed effects 38 One of the most important aspects when applying linear regression methods is the selection of independent variables. Including many independent variables has both advantages and disadvantages. On the positive side, adding more variables can increase the explanatory power of the model, allowing it to account for a greater proportion of the variability in the dependent variable. Additionally, incorporating relevant variables can reduce bias in the coefficient estimates by accounting for more influences on the dependent variable. Moreover, a model with more variables can capture complex interactions between different factors, offering a richer perspective on the relationships between variables. There are also several drawbacks about including an excess of variables. One major issue is the risk of overfitting, where the model becomes too closely tailored to the training data and captures noise rather than underlying patterns. Another problem is multicollinearity, which arises when independent variables are highly correlated with each other, making it difficult to determine the individual effect of each variable. This can inflate the standard errors of the coefficients and undermine the reliability of the results. Additionally, models with many variables can become complex and difficult to interpret. We have run some models with different variables until we have reached the final variables to be included. Just as an example, we thought about the monthly number of flights that arrive to Singapore as an explanatory variable for the number of visitor arrivals. When we include this variable and we plot the correlation matrix, we see that it is very correlated with the dependent variable (0.95). For this reason, we have not included the number of flights in the final model, as it would present multi-collinearity issues. Figure 16: Correlation matrix 39 This correlation is an intuitive result, because the more flights, the more visitors capacity that Singapore can welcome. To see this relation visually, we can plot a joint graph. Figure 17: Monthly comparison of visitor arrivals vs number of flights 40 4.5 Interpretation of results After completing the analysis for the Singaporean Formula One Grand Prix, we now turn to interpreting the results, with a particular focus on the robustness and validity of our model. Reviewing all the residual plots and correlation matrices, we can confirm that our model meets the OLS assumptions, including homoskedasticity of the residuals, linearity in the coefficients, absence of autocorrelation in the residuals, and no multicollinearity among the variables. These verifications lend credibility to our findings and validate the outcomes of our model. The most important variable in our model is the coefficient related to the F1_Event variable, which means that our coefficient of interest is β1. Given that the dependent variable has a logarithmic form, this coefficient represents the percentual impact of hosting the Singaporean F1 Grand Prix on the dependent variable. A positive coefficient would mean that hosting the event increases the number of visitor arrivals or the tourism receipts during the month of the event. On the contrary, a negative coefficient would mean that hosting the F1 event has a negative impact on the tourism indicators. By construction, the multiple regression model solved by OLS uses a t-statistic test to determine if each coefficient is different from 0 or not. Therefore, the null hypothesis for each coefficient is that the coefficient equal 0 and the alternative hypothesis is that it is different from 0. If we obtain a p-value higher than the significance level of 5% (0.05), we can reject the null hypothesis. Otherwise, if the p-value is higher than the significance level, we cannot reject it and so we assume that the coefficient is not statistically different from 0. For each of the two dependent variables (visitor arrivals and tourism receipts), we have constructed different models. More precisely, it is one base model and four modifications adding new control variables for year fixed effects and for month fixed effects. In all our models, the coefficient β1 has an associated p-value higher than the significance level of 0.05. As a consequence, we can interpret that the coefficient is not statistically different from 0 even if we control for the economic and time fixed effects. This is the main interpretation of the model. 41 Hosting the Singaporean Formula One Grand Prix does not have a significant effect on the number of visitor arrivals or the amount of tourism receipts. 5 Conclusions After conducting a thorough review of existing literature and delving deeper into the case of the Singaporean Formula One Grand Prix, we can pinpoint a key conclusion from our paper. Economic impacts of hosting a sporting mega-event can vary significantly across different events. This variance can be influenced by numerous factors such as the type of event, the economic conditions of the host country, the ability to develop sustainable infrastructure, and the inherent tourism appeal of the region. Generally, the long-term benefits of such events surpass the short-term gains, primarily due to the substantial initial investments required. Hosting mega-events often results in a legacy that justifies the bid for their organization, despite the potential for financial losses. Specifically, our focus was on how Formula One events affect regional tourism metrics. Previous research suggests that these events do not yield positive effects in Europe and Australia, which shaped the hypotheses for our model. These posited that hosting a Formula One event significantly affects tourism arrivals or receipts in the region. Our analysis of the Singaporean Grand Prix data do not support these hypotheses. Using multiple linear regression models, we found no significant statistical correlation between hosting the Formula One event and changes in tourism indicators, so we can reject the hypothesis. The implications of our findings suggest that public authorities, politicians and other stakeholders should reconsider that organizing a Formula One Grand Prix Table 4: Summary of the β1 coefficients and their associated p-values