1
The economic impact of mega sport
events on the hosting country
A cross country analyses of the impact on GDP growth and Unemployment rate
Hugo Zwetsloot 02-02-2016
10355359 University of Amsterdam Macroeconomics / International economics Ron van Maurik
Economics and Business economics
Abstract
This thesis analyses whether there is a relationship of hosting a mega sport event on the annual growth of GDP and Unemployment rate. Here it will be researched throughout a panel data regression on 17 countries which have hosted a World Cup football, the Summer
Olympics, the Winter Olympics or the European Championship of football. A regression analyse will be conducted and this will show if there is a significant impact of a mega sport event.
2
Table of Content
1. Introduction ...3 2. Theoretic background...4 3. Methodology...9 4. Results...13 5. Conclusion...15 6. Appendix……….17 7. Bibliography ...213
1. Introduction
On October the 30th in 2007 the FIFA, Federation Internationale de Football Association, decided to host the World cup football 2014 in Brazil. Furthermore, on 17th of May the bidding process started of who will host the next Olympic games of 2016. Four candidates remand in the race for this precious event: Chicago, Madrid, Rio de Janeiro and Tokyo. Eventually at the final voting on 2 October 2009 Brazil (Rio de Janeiro) came out as the winner and will host the Olympic games of 2016.
With millions of spectators and billions of people who watch the event live through television, clearly such an event can be titled as a mega event. Hosting an event of this size gives a country the opportunity to show the rest of the world how attractive they are on political, economic and cultural area. With no doubt hosting a mega event will result in an enormous international publicity and could put a country back on the map (Matheson & Baade, 2004).
The main discussion is still going on if these mega sport events bring prosperity to a country or result in an economic downfall. Matheson (2006) states that in previous mega sports events like with the National football league (NFL) the Super Bowl only already generated an economic impact of around 300 to 400 million. With the World Series of basketball Matheson claim that this will make an economic benefit of 30 to 110 million. Organizing a mega event like the World cup football or the Olympic Games believe to produce an even bigger economic effect.
On the other hand, Humphreys and Prokopowicz (2007) argue that through a simple cost benefit analysis the cost of hosting such an event will exceed the economic impact. They believe that the extra income from tourism spending’s will not cover the costs. They do state that if there would be a positive outcome from the hosting of a mega event it would come through other factors such as the improvement of the infrastructure.
From the previous studies there has not been a definite answer to whether a mega sport leads to positive economic impact. In this research a multiple regression analyse will be conducted to get a closer answer for the ongoing discussion. The following research question will be as followed:
4
This question will be answered by collecting data from countries who have hosted a World Cup football, Olympic Games or the European Championship. With the use of panel data there will be made a multiple regression analyse to see if a mega sport event as a significant effect on the annual GDP growth rate and the Unemployment rate of the hosting country. This research is different than previous literature because no research yet has included the World Cup football, the Olympics and the European championships in one regression. By including these mega events in one regression it will give the ability to include more countries then other literature has done.
This paper will start with an overview of the previous literature about the economic impacts of mega events. Secondly, the methodology of the research method will be discussed. Afterwards, the result will be interpreted and as last a conclusion of these result will be done.
2. Theoretic background
In this paragraph the work is put together of the previous literature which have led to
different analyses. In the past literature there are several ways how these analyses have been conducted. In this paragraph there will be the focus on which researches have been done and what the outcome was from these analyses.
To measure the effect of a mega sport most governments use the ex-ante or the ex post analyses (Matheson & Baade, 2006). In the ex-ante analyses a prediction will be done to estimate the economic impact of a mega sport event. These predictions are mostly done for the prediction of a single variable for example tourism. In an ex ante analyse the amount of spectators that the event will attract, how long the visitors will stay and how much they will spend will be predicted. These predictions are the direct effect of a mega event. To measure the indirect effect the ex-ante study makes use of the multiplier effect (Matheson & Baade, 2006). The multiplier analyses is used to estimate the impact of a particular inflow of money in an economy. The multiplier predicts that an input of money will result in a chain of spending’s. If this approach is used it is of significant importance to estimate this multiplier correct, if this is off by a little it will lead to wrong estimates throughout the secondary impact of the multiplier (Barclay, 2009). There are three main multipliers used in making predictions which are the sales multiplier, household income multiplier and the employment multiplier. Out of this the employment multiplier is the hardest to estimate correct and could
5
lead to that the predictions are often estimated to high (Kasimati, 2003). Based on the ex-ante analyses many predictions have been done and this method is by far still the most used method to estimate the effect of a World Cup or the Olympic games.
Most of the ex-ante analyses give high expectation of a mega event by showing high positive outcomes. Kasimati (2003) states that most of the ex-ante analyses are possibly biased due to the fact that commissions who conduct these studies are in favour of hosting such a mega event. Matheson & Baade (2003) also put their questions on some of these prediction. Matheson and Baade have done research on this matter throughout studying the employment data from metropolitan area of the summer Olympics in 1984 and 1996. They came to a result that the Olympics in Los Angeles of 1984 created an unexplained increase of 5000 jobs. This increase of jobs would produce a boost for the economy of 300 million. For the Atlanta Olympics Matheson & Baade (2003) studied the period from 1994 till 1996. They saw that the predicted amount of 77,000 new jobs was overestimated and they came to the conclusion that the job increase would be between 3,500 and 42,000 new jobs. If you would take this number generously it would nearly be half of what was predicted (Matheson & Baade, 2003).
Késenne (2006) points out a second effect which leads to an overestimation of the predictions. Késenne states that during a mega sport event the crowding out effect occurs. This means for example the new labour needed for the extra construction activities for the event will be pulled out of other branches/companies or even from outside the country. This will reduce the net positive effect for the local economy. Moreover, crowding out also occurs with the government spending’s on the mega event. The government could have spent their money on other projects where no crowding out effects would occur. The crowding out effects are because of this to be considered as part of the opportunity costs. Matheson and Baade (2004) add an extra crowding out effect. This relates to the extra tourism attracted to the mega event. A mega event will create an extra flow tourism but will possibly reduce the tourism which come every year to this country. During the World cup of 2002 in Japan and South Korea the increasing numbers of European visitors due to the World cup resulted in a similar decrease in usual tourism and business visitors (Matheson & Baade, 2004).
An ex post study leads to a more reliable kind of information. In an ex post study the host city or country will be examined on the economic situation before after the mega event. It will be made sure that the economic influences contributed beside the mega event are isolated. Another way of examining the ex post economic impact is comparing it to a country or city with a comparable economic situation where the mega event was not hosted
6
(Kasimati, 2003). In former ex post studies the emphasis was mainly on retrieving data about personal income, per capita income, taxable sales and employment.
Malfas, Theodoraki & Houlihan (2003) have done ex post research on the effect of employment and came to the conclusion that a mega event does create a significant amount of new jobs. This job creation does not have to be a direct result of the mega event but could also come from the tourism industry, the retail industry and the construction industry. The new employment is needed due to the increase in visitors and the development of the infrastructure. One of their result was of the 1992 Olympics in Barcelona, of a period between 1986 till 1992 the unemployment rate decreased from 18.4% to 9.6% (Malfas, Theodoraki & Houlihan, 2003).
Another post effect of a mega event is the increase in tourism. England experienced this during the 1996 European Championship of football. The tourism boost resulted in its first positive trade balance since the beginning of 1995. The European Championship of football attracted 280.000 visitors with in total a spending of 120 million over the eight cities where was played. In the same way during the 1998 World cup in France it attracted over 10 million visitors and resulted in more than half a billion of spending’s on food, hotels and travel (Malfas, Theodoraki & Houlihan, 2003).
The fundamentals for both ex ante and ex post analyses are from two different types of models, the input–output (I-O) and the computable general equilibrium (CGE) model. The input-output is a widely used model in economics to measure the economic impact and is established by Leontief in 1940 (Kasimati, 2003). The input-output model uses input-output tables to represent the interactions between different parts of the national economy on the industrial sectors to estimate the economic impact. The model makes use of production or price categories. The output of one industry will be the input of another industry. A limitation of this model is that is very depended on the availability of these input-output tables.
Furthermore a shortcoming of this model is that it does not make use of economies of scale, production close to full capacity and price changes to the demand shifts. This is also one of the reasons why many ex ante analyses are overestimated (Kasimati, 2003).
Because of the shortcomings of the I-O model Sydney used, during their Olympics of 2000, the computable general equilibrium (CGE) to measure the economic impact this mega event. The CGE is a new developed model which is getting more and more popular (Madden, 2006). The CGE model is a framework of the economy and uses aspects of the input-output model. The difference with the I-O model is that it makes use of price changes and includes economy constrains. Because it incorporates prices changes it is very useful during a mega
7
event. It can automatically handle shifts in the economy which will happen often during a mega event (Madden, 2006). The I-O model lacks this ability and would overlook these shifts. Furthermore the CGE model applies flexible prices so it can move to adapt to the market. This makes the model significantly more accurate (Kasimati, 2003).
Studies which have researched the economic effect on econometric basis of previous mega events came to different conclusion then of the ex-ante predictions. Maennig and Allmers (2009) have done empirical research of the economic effect of het World Cup of 1998, 2006 and 2010 on overnights stays and retail sales. With ex ante and ex post analyses the conclusion was that a mega event does have an impact on tourism. Maennig and Allmers (2009) found that tourisms has no to little significant impact on the hosting country. They even noticed a decline in tourism in the month July when France hosted the World cup in 1998. They relate these finding with the theory about the crowding out effect. A World Cup cause that the regular tourism flow will postpone or even cancel their vacation due to the fact that a World Cup is hosted. The findings for the retail sales also lead to contradicting result compared to what would be expected during a World Cup. The retail sales had no significant effect and, even though it is not significant, they found a decrease in retail sales during the World Cup period. The theory about why a World Cup leads to a decrease in retail sales could be linked to the “couch potato effect”. Consumers might shift from their normal
consumption because of the World Cup. They choose to change their consumption behaviour to watch the games at home and consume themselves with “potato” fast food (Maennig & Allmers, 2009).
The economic impact is very difficult to measure mainly because the effect is “intangible”. Hosting a World Cup gives the country a great opportunity to present
themselves to the world and show their beauty. One way to capture these “intangible” effects is through the Anholt Nation Brans Index (BSI) (Maennig & Allmers, 2009). The BSI
measures quarterly the brand image of a country on cultural, political, commercial, human assets, investment potential and tourist appeal. Germany experienced a rise in this ranking system after hosting the World Cup in 2006. Germany climbed from the 5th place to the 2de place. This result could suggest that a mega event does have a great impact on a positive brand image awareness. Another factor which will be influenced is the happiness of the citizens during a mega event, the so called “feel good effect”. Limited research has been done but Kavetsos and Szymanski have studies this effect on the European country over the past 30 years. There result was that there is a positive significant impact of this during a mega event (Kavetsos & Szymanski, 2008).
8
Besides the results not being significant from zero Maennig and Allmers (2009) state that the effect of a World Cup could be significant for a developing country as South Africa. The reasons they give are that the “couch potato effect” is less likely to happen because of the cultural difference in going out by the football fans. The spending’s might become higher instead of lower because of this cultural difference. Furthermore the negative crowding out effect might not occur. The World Cup would be hosted in their winter period and this is at the same time their low season of tourism (Maennig & Allmers, 2009).
The majority of the ex post analyses on econometric basis suggest that on regional income or employment there is no till little impact of a mega event. Hagn and Maennig (2008) have done research on the employment effect during the World Cup of 1974 in Germany. They computed a model with dummy variables for every city where a game was played. All of these dummy variables turned out to be not significant different from zero. This would suggest that there is no significant impact on employment during the World Cup of 1974. Matheson & Baade (2004) conclusion on the effect on regional income due to a mega event resulted in a similar outcome. A study which did result in a significant impact was by Tien, Lo & Lin. They studied the relationship between the Olympic Games and the GDP, unemployment and investments. They collected data roughly about the period 1972 till 2007, these dates vary in their research since missing data occurred. In their research they used panel data over 15 different countries and created dummy variables for three periods: before, after and during the Olympic Games were hosted. Their results were that only the period before the Olympic Games turned out to be significant (Tine, Lo & Lin, 2011). This varies from most empirical research where here a mega event does not lead to a significant impact. The dummy variable was in all three regression significant for the pre-game phase on GDP, unemployment and investments. Tien, Lo & Lin (2011) come to the conclusion that hosting the Olympics will generate only a significant impact on short term effect.
To sum up, the outcome from ex ante predictions throughout the input-out model or the computable general equilibrium often leads to overestimation of the positive impact. Studies which conducted an empirical research mostly concluded that there is no effect found on the tourism arrivals, retail sales, GDP, employment or investments. Most studies resulted that there is no significant impact on these variables during a mega event. This is contracting to what ex ante predictions give. In the following paragraphs my research will be discussed.
9
3. Methodology
In this chapter the method of research will be described and the model will be explained. This research investigates the economic impact of a mega event on the hosting country. First of all the variables used in the regression will be explained. Secondly we will explained which and why these particular countries are chosen and as last the model will be described.
3.1 Data
To analyse the economic effect of a mega event on a hosting country we will run two regression to analyses the on the change in Gross Domestic Product Growth (annual %) and Unemployment Rate. Furthermore Foreign Direct Investments (% of GDP), Current account balance (% of GDP) and the government expenditure (% of GDP) will be the independent variables. These variables are chosen because during a mega event these in my opinion will be most influenced by a mega event.
The dependant variable GDP is referred to the GDP per capita. The GDP per capita is calculated by the gross domestic product divided by the midyear population. GDP contains the sum of all the gross value contributed by the citizens in the economy. Here it is still needed to add the product taxes and detract the subsidies. GDP is a good dependant variable since it is a good and frequently used measurement to measure the economic performance of a country. We use the annual GDP growth to calculate the growth difference by year. The Unemployment rate is a percentage of the total labour force divided by the share of labour force who is without a job and is seeking for a job. In this study the Unemployment rate will used as dependant variable as well as an independent variable to estimate the annual GDP growth.
The FDI (Foreign Direct Investments) refers to the direct invested equity flow from the particular economy. The direct invested equity includes reinvested earnings and capital. Direct investments are a cross country investment with on economy having a significant degree of power on the management of another company which is located in a different economy. To classify as a direct investment relationship it should possess 10 percent of the shares. FDI will be used as an independent variable as a percentage of the total GDP.
10
Government expenditure is measured by all the government expenditures on the purchases of goods and services. Government expenditure will be taken as a percentage of the total GDP. All the data used is in current US dollars.
In this research seventeen different countries are used to examine whether there is a GDP effect and an Unemployment effect during a period of a mega event. These countries have all hosted an Olympic, a World Cup, or the European Championship Football. I have picked the following countries: Italy (World cup 1990, Winter Olympics 2006), Spain (Olympics 1992), USA (World cup 1994, Summer Olympics 1996, Winter Olympics 2002), France (World cup 1998, Winter Olympics 1992), Australia (Summer Olympics 2000), Greece (Summer Olympics 2004), Germany (World cup 2006, European championship 1998), China (Summer Olympics 2008), South Africa (World cup 2010), Brazil (World cup 2014), UK (European championship 1996,Summer Olympics 2012), Portugal (European championship, 2004), Canada (Winter Olympics 1988, Winter Olympics 2002), Japan
(Winter Olympics 1998, World cup 2002), South Korea (Summer Olympics 1988, World cup 2002). During the period of 1985 till 2014 data is collected for all the variables used in the dataset. This time frame has been picked because before 1985 it became more difficult to find enough data. The data used for this research is mostly gathered from the World Bank, OECD and the IMF. The missing data for some countries have been gathered from their national bank sites. For China it was needed to collect a part of the data through the State
Administration of Foreign Exchange. The final sample contains 510 data points and 17 countries. With this dataset it is possible to conduct a panel data regression.
3.2 The Model
Before analysing the outcome of the data it is needed to conduct multiple test to check how the data is fitted. As first it is needed to test for stationarity. This is important because if the variables are stationary it means that variance, mean and autocorrelation does not chance over time and do not follow any trends. By conducting the Dickey–Fuller test and the Levin–Lin– Chu test for stationarity it resulted that all of the variables are stationary for a 0.000 p-value. All of the variables are stationary at a 1%, 5%, 10% level. Next it was tested if the dataset had autocorrelation. By conducting the Wooldridge test it is possible to see if there is autocorrelation. This resulted in a P-value of 0.0005, with the H0 to be that there is no autocorrelation. This means that there is autocorrelation in the data set over time. This could make sense since a year where there is positive GDP growth it is mostly followed up with
11
positive GDP growth the next year. Before it can be tested if the variables are not constant, so if there is heteroscedasticity, it is important to know for panel data if to use a random effect regression or a fixed effect regression. To decide between fixed or random effects a Hausman test is used where the null hypothesis is that the preferred model is random effects vs. the alternative the fixed effects. It basically tests whether the unique errors are correlated with the regressors, the null hypothesis is they are not. The Hausman test resulted in a P-value of -3.91. From this outcome it means it is important to use the random effect regression. The Breusch and Pagan Lagrangian multiplier test for random effects confirms this also. This test examines if it is better to do an OLS regression instead of the random effect regression. The P-value is 0.000, with the H0 to do an OLS regression. So it is from great importance to run a random effect regression. As last to test for heteroscedasticity the LR test is the only
possibility for a random effect regression. This outcome leads to P-value of 0.000 and means that there is heteroscedasticity in the dataset.
This thesis helps to analyse if there is a significant impact of a mega sport event. To research this the following models has been computed:
(1) GDP (Annual Growth) = β0 + β1*Unemployment Rate + β2*FDI (% of GDP) + β3*CA (% of GDP) + β4*Gexp (% of GDP) + β5*PreEvent + β6*Event + ε
(2) Unemployment rate = β0 + β1* Unemployment Rate, t-1 + β2*FDI (% of GDP) + β3*CA (% of GDP) + β4*Gexp (% of GDP) + β5*PreEvent + β6*Event + ε
In equation 1, GDP is the dependent variable and Unemployment rate, FDI, Current account balance (CA), Government expenditure (Gexp) and two dummy variables the independent variables. For equation 2, the Unemployment rate is the dependant variable and
Unemployment rate at time t-1, FDI, Current account balance (CA), Government expenditure (Gexp), the PreEvent dummy and Event dummy the independent variables. The lagged term for unemployment is introduced to correct for possible serial correlation. Furthermore in previous regressions on unemployment one or more lagged terms have been included in the models. GDP has not been used because it would result for simulations causality. In both regression the dummy variables are created to see if there is a significant impact of a mega sport event. The PreEvent dummy will be 1 in the five years before the mega event will hosted. This has been done since the effect of a mega event probably will start already when
12
the preparations are started. For example building more infrastructure will stimulate the employment. The dummy for Event will be 1 in the year when the mega event is taking place. If one or both of these dummies are significant we can conclude that there is indeed a
(positive) impact of World Cup, Olympics or the European championships. If this is the case it is beneficial for a country to hosts such a mega event.
13
4. Results
In this paragraph the result of the empirical research will be discussed. Based on the model which was discussed previously we can interpret these result. First the result from equation 1 will be discussed and after the result from equation 2, the regression on the effect on
Unemployment rate.
Now that the dataset has been tested on multiple criteria to make sure to do the right regression the following coefficients have been formulated:
Equation (1) (2)
Dependant variables GDP Unemployment Rate
Independent variables Constant 15.17796 *** (10.61) -1.444918 * (-1.94) GDP … …
Unemployment rate (% of total labor force)
-0.0899916 (-1.57)
…
Unemployment rate, t-1 (% of total labor force)
… 0.8712053 ***
(13.06) Foreign Direct Investments(% of
GDP)
0.1213125 ** (2.20)
-0.0856408 ** (-2.30)
Current Acount Balance (% of GDP) -0.0455865 (-1.48) -0.0203248 (-1.40) Government expenditure (% of GDP) -0.6420793 *** (-8.26) -0.1466089 *** (2.81) PreEvent 0.2712758 (0.87) -0.0766776 (-0.40) EventYear 0.591679 (1.18) -0.0092064 (-1.94) R^2 0.2938 0.8590 N 510 510
14
Following the results of equation 1 it can be said that the unemployment rate as barely to no effect on the GDP growth. The unemployment rate has at a 10% level not significant impact. This could be traced back to the literature where multiple articles state the crowding out effect. The extra employment needed during a mega sport event mostly doesn’t solve the unemployment but instead the extra employment will be pulled away from other industries or even from abroad. The independent variable FDI does have significant impact on the GDP growth. The variable is significant at a 1% level. Economic studies state that during a mega event the FDI and GDP will be significantly affected. This makes sense since in economic theory when more investments are done from foreign investors this mostly leads to GDP growth. The effect of the current account balance is however not significant. The current account balance has a negative coefficient which means that when CA increases with 1 the GDP growth will decrease with -0.05. The current account doesn’t give a large change in the GDP growth and this could be due to the fact that the impact also could rely on the capital accounts balance. Generally in economic theory it is expected when a country has a current account surplus this should be beneficial for the economy and so it could lead to GDP growth. Government expenditure (Gexp) also has a negative relationship with the GDP growth. The Government expenditure does have a significant impact on the GDP growth. An increase in the government expenditure leads to a decrease in the GDP growth. Now to be able to conclude that a mega sport event could have an economic impact the two dummies were created. In this regression both the dummy variables have positive coefficient but both variables are not significant at a 10% level. This means that the five years of preparation before a mega event does not give a significant impact on the GDP growth. The same analyse will hold for the year when the mega event is hosted in a particular country. These results are in line with the findings of other empirical results. In most scholars there was no effect to be found during or before the hosting of a mega event. The reasons for this outcome can have several explanations. In the data the impact of a mega event on a large country will probably be less significant when it would be hosted by a smaller country. Since most countries are from a big economies it is harder to prove that there is a significant effect. Furthermore hosting the Olympic Games only take place in one city, so the impact could be there for this particular city but for the country as a total the impact could be significant less. As last it should be said that the impact for a developing country could have a larger effect. Since most mega events are hosted by a developed country this could be another explanation to the not being significance of the dummy variables.
15
event on the Unemployment rate. The PreEvent dummy and the Event dummy are both not significant at a 10% level. This means that the effect of a mega event at the preparations and the year of the mega event does not lead to a significant impact on the unemployment rate. This is in line with the previous literature. Maennig and Hagn studied the employment effect of the World cup 1974 and found that all their created dummies where not significant. Also the study of Tien, Lo & Lin concluded that there is a weak to no impact of the Olympics on the unemployment rate. The other independent variables do result in a significant effect on the unemployment, except for the current account balance. This indicates that FDI and Government expenditure do have a significant effect on the unemployment rate.
5. Conclusion
For this research the following research question has been analysed: Do mega sport events
have a positive economic impact on the hosting country? Throughout reviewing past
literature and conducting a panel data regression on the collected data of 17 countries it should give an answer to this question.
In the existing literature there was not a clear answer given to this question. Many articles state that mega events lead to positive economic impact. However the literature is critical about these analyses which are mostly done with ex ante predictions. They state that the ex-ante predictions are often exaggerated due to the fact that that the predictions which are conducted by parties who are in favour of hosting the mega events in their country. In the literature on previous econometric analyses mostly result that there is no significant effect to be found during a mega sport event. The panel data regression of Tien, Lo and Lin did result in an effect for the period before the Olympic Games. However this effect is therefor only to be proved to be a short term effect and so in the long term the Olympics still does not lead to a significant effect. Studies do indicate that the impact of mega sport event mainly improves the on brand awareness for the hosting country. Furthermore it will result to a feel good effect.
Based on the regression results we can conclude that hosting a mega event does not contribute to a significant impact on the annual GDP growth and the Unemployment rate. The dummy variables in both regression resulted to be not significant. This answers the research question that a mega sport event does not result in a significant economic impact for
16
the hosting country.
To be able to get an unambiguous answer for this ongoing discussion more research should be done on this matter. Due to the fact that most mega sport events are hosted in a developed nation and furthermore are from big economies it makes it more difficult to find a significant impact. However there is a tendency for developing nations to host such events. In the future it will be possible to study the impact for devolving countries or smaller economies to result in a significant impact. I believe that a regression analyses similar to this one should give a more clear answer when more data will be available. More variables should be put in the model and furthermore it will be needed to collect more data over time and include more countries.
17
6. Appendix
Besides the results of the dummy variables it is still interesting to see what the impact looks like in a descriptive analyses:
In the graphs with the abbreviation WC it is meant that in this year a World Cup football has been organized, SO stands for the Summer Olympics, WO for the Winter Olympics and EC for the European Championship football. The USA is one of biggest economies and countries in this research. What is noticeable is that in the graph, about GDP annual growth rates, that during periods when a mega event is hosted in the USA it constantly had a positive increased GDP growth. The periods before hosting the World cup in 1994 have been of decreasing GDP growths. But the year when the mega events were hosted it did result in a significantly positive reaction.
-4 -3 -2 -1 0 1 2 3 4 5 6 1980 1985 1990 1995 2000 2005 2010 2015 2020 GD P An nu al G ro wt h Years
USA
WC SO WO18
The same effect has been observed for South Africa, but then more heavily. South Africa is in the empirical research one of the more developing countries and in the year of the World Cup in 2010 it resulted in a 4.5 percent point difference from the year before. The GDP growth has declined significantly in the years before the World Cup. This could be a result of the extremely high costs which are needed to meet the requirements of the FIFA. South Africa spend over 3 billion dollars for the preparations of the World Cup and 1.1 billion alone was for building and renovating the stadiums.
Japan is the first country to host a World Cup in Asia. During a period of 1998 till 2002 Japan has organized the Winter Olympics in 1998 and the World Cup in 2002. The World Cup was hosted together with South Korea at that time. Japan experienced during this period a period of chancing GDP growth or decline. Both after the Winter Olympics and the World
-3 -2 -1 0 1 2 3 4 5 6 7 1980 1985 1990 1995 2000 2005 2010 2015 2020 GD P An nu al G ro wt h Year
South Africa
WC -8 -6 -4 -2 0 2 4 6 8 1980 1985 1990 1995 2000 2005 2010 2015 2020 GD P An nu al G ro wt h YearJapan
WO WC19
Cup Japan experienced a period of GDP growth till the economic crisis in 2008 started. But real impact of the two mega events is not discernible. Japan is still coping with economic downfall so it could be that this reduces a possible effect of the mega events.
Germany experienced in the two different periods when Germany hosted a mega event a kind of similar effect. In 1988 when (west) Germany hosted the European championships there was an increase in the GDP annual growth. With almost a peak at when the European
championships where hosted. Similar at the World Cup of 2006 the GDP growth is at its peak for this period in time. The years before it is noticeable that Germany increased there GDP growth significantly.
Greece, with one of the smallest economies in this research, experienced in the year around the Summer Olympics of 2004 a period of constant positive GDP growth. The period before the Olympics where with increasing GDP growth till almost at its high point when the
-8 -6 -4 -2 0 2 4 6 8 1980 1985 1990 1995 2000 2005 2010 2015 2020
Germany
WC EC -10 -8 -6 -4 -2 0 2 4 6 8 1980 1985 1990 1995 2000 2005 2010 2015 2020 GD P An nu al G ro wt h YearGreece
SO20
Olympics where hosted. Followed after a period of decreasing GDP growth to a GDP decline when the economic crisis started in 2008.
21
7. Bibliography
Baade, A.R., Baumann, R., Matheson, V.A. (2008). Selling the Game: Estimating the Economic Impact of Professional Sports through Taxable Sales. Southern Economic
Journal 74 (3): 794-810
Barclay, J. (2009). PREDICTING THE COSTS AND BENEFITS OF MEGA-SPORTING EVENTS: MISJUDGEMENT OF OLYMPIC PROPORTIONS Economic affairs. 62-66
Humphreys, B.R., Prokopowicx, S. Assessing the impact of sports mega-events in transition economies: EURO 2012 in Poland and Ukraine. International Journal of Sport
Management and Marketing. 5 (2)
Kasimati, E. (2003). Economic aspects and the Summer Olympics: a review of related research. International Journal of Tourism Research, 433-444.
Kavetsos, G., Szymanski, S. (2008). The Impact of Mega Sporting Events on Happiness.
Hamburg Symposium of Sport and Economics.
Késenne, S. (2005). Do We Need an Economic Impact Study or a CostBenefit Analysis of a Sports Event? European Sport Management Quarterly, 5:2, 133-142,
DOI: 10.1080/16184740500188789
Madden, R.J. (2006). ECONOMIC AND FISCAL IMPACTS OF MEGA SPORTING EVENTS: A GENERAL EQUILIBRIUM ASSESSMENT. Public Finance and
Management 6 (3): 346-394
Maennig, W., Allmers, S. (2006) Economic Impacts of the FIFA Soccer World Cups in France 1998, Germany 2006, Andoutlook for South Africa 2010. Eastern Economic
Journal. 35 (4): 500-519
Maennig, W., & Hagn, F. (2008). Employment effects of the Football World Cup 1974 in Germany, Labour Economics 15: 1062-1075.
Matheson, V.A. & Baade, R.A. (2004). Mega‐Sporting Events in Developing Nations: Playing the way to prosperity. South‐African Journal of Economics 72, (5): 1085‐ 1096.
Matheson, V.A .(2006). "Mega-Events: The effect of the world’s biggest sporting events on local, regional, and national economies". Economics Department Working
Papers. Paper 68. http://crossworks.holycross.edu/econ_working_papers/68
Malfas, M., Theodoraki, E., Houlihan, B. (2004). ‘Impacts of the olympic games as megaevents’, Municipal Engineer 157(3): 209–220.
22
Tien, C., & Lo, H., & Lin,W. (2011). The Economic Benefits of Mega Events: A Myth or a Reality? A Longitudinal Study on the Olympic Games, Journal of sport management 25: 11-23.
