Skip to main content
SpringerLink
Log in

Market impact of international sporting and cultural events

  • Published:
Journal of Economics and Finance Aims and scope Submit manuscript

Abstract

We study market reaction to the announcements of the selected country hosting the Summer and Winter Olympic Games, the World Football Cup, the European Football Cup and World and Specialized Exhibitions. We generalize previous results analyzing a large number and different types of mega-events, evaluate the effects for winning and losing countries, investigate the determinants of the observed market reaction and control for the ex ante probability of a country being a successful bidder. Average abnormal returns measured at the announcement date and around the event are not significantly different from zero. Further, we find no evidence supporting that industries, that a priori were more likely to extract direct benefits from the event, observe positive significant effects. Yet, when we control for anticipation, the stock price reactions around the announcements are significant.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. For example, Germany spent over 1,4 billion euros building or rehabilitating 12 stadiums for the 2006 Soccer World Cup of which 35% were funded by taxes and Greece spent over 1 billion euros in the 2004 Olympic Summer Games on security (cited by Matheson 2006).

  2. Dwyer et al. (2005) propose a more comprehensive computable generating equilibrium model (CGE) to assess the economic impact of such events.

  3. The benefits may not go all to the host country but to foreign neighbor countries or multinationals. For example, the 16 official partners for the 2006 FIFA World Cup in Germany were multinationals and only two were German. Yet all the official suppliers were German companies.

  4. Public funding is often required for the event infrastructures. This could imply that other potential more productive investments are not funded (or postponed) or that taxes have to rise (Siegfried and Zimbalist 2000). Those public expenditures may or not have positive impact on the economy. Sporting events specialized infrastructures such as stadiums or swimming pools have a limited use and potentially benefit only a small part of the tax payers that paid for it. More general construction projects such as cities core redevelopment and infrastructure building may benefit more directly the country or local community. Similarly, the benefits from investing in cultural or lifestyle amenities can attract highly educated and creative young people that are essential to economic growth.

  5. The authors suggest that host cities accumulated losses of US$ 5.5 to US$9.3 billion as opposed to the US$4 billion gain estimated by the organizers.

  6. Please refer to Matheson (2006) for an extensive survey of ex ante and ex post economic impact studies (Tables 1 and 2).

  7. For thorough reviews, see, for example, Fama (1991) and Dimson and Mussavian (1998). Several recent studies present evidence contrary to market efficiency suggesting either overshooting in prices or gradual information dissemination (see for example, Fama 1998).

  8. See, for example, Woolridge and Snow (1990), Jones et al. (2004) and Titman et al. (2004) and references therein.

  9. Related with this is what is sometimes designated by Capital Myopia that refers to excessive investments pursued by companies that erroneously believe that there is scope for further profitable capital investments ignoring that competition will drive away economic rents.

  10. See, for example, Acharya (1993) or Akhigbe et al. (2004) and the references therein.

  11. A similar study conducted by Berman et al. (2000) found no significance effect on the overall market, and only limited effects on stock prices of infrastructure development companies based in New South Wales where the Olympic Games were hosted.

  12. The statistical null hypothesis tested in Section 5 is that the impact of the announcement of the event is null and so forth for the other hypotheses.

  13. In alternative, positive or negative market reactions, respectively for the winning or losing countries could reflect investor sentiment. A stronger (negative) effect for losing countries is consistent with behavioral arguments.

  14. Some events are co-organized by two or more countries. For example, Belgium and Netherlands organized the 2000 European Football Cup together. In that case we consider them as separate observations.

  15. There are a few cases for which there is no market information for the winning country when the nomination was announced. For example, this is true for the 1988 Summer Olympic Games in the former Soviet-Union or the 1988 Winter Olympic Games in Korea.

  16. Stock market information was not available for several losing countries by the time of the nomination announcement (for example, China, in respect to the 2000 Summer Olympic Games or Morocco, in respect to the 1998 and 2002 World Football Cups).

  17. Datastream indices were preferred over other domestic market and industry indices when available because they are constructed on a uniform basis across markets and are not backfilled with firms added or deleted from the index. The exception was the total return series for Spain general index (IBEX) obtained directly from Bolsa de Madrid.

  18. Kothari and Warner (2006) show that the tests are not highly sensitive to the benchmark model of abnormal returns. Market-adjusted returns are not included here for all tests. Results are available upon request.

  19. Because we use continuously compounded returns, buy and hold returns for a specific time-span are achieved simply summing the log returns. If we assume that discrete returns are distributed as iid log normal variables, cumulative log returns are normal distributed.

  20. The benchmark to compute industry market- and risk-adjusted abnormal returns was the country’s total return market index.

  21. Kothari and Warner (2006) show that with short horizons, the usual test statistic is not highly sensitive to assumptions about the cross-sectional or times-series dependence or normality of returns. Further, they show that short horizon event study tests are generally well-specified but the power of the tests is sensitive to sample size and firm characteristics (such as volatility). For firms with low volatility, sample size of 20 is enough attain full power for a 1% abnormal return.

  22. The tables below report the statistics for the usual Brown and Warner (1980, 1985) parametric tests and for the sign test. Other results are available upon request.

  23. We run the same regression for market CARs (instead of industry CARs) using OLS (instead of fixed effects) without the industry-specific variable REP.

  24. Veraros et al. (2004) argue that the difference in the reaction of Athens and Milan stock exchanges could result from economy size differences (Greece and Italy) and the importance of the two cities potentially hosting the event (Athens and Rome).

  25. If stock prices are influenced by sentiment, δ reflects the effect of positive sentiment while α captures the negative sentiment in prices. Further, if sentiment effects are more pronounced for losing countries, |α| > |δ|. Finally, ϕ = 0, reflecting that prices are affected by investor sentiment, regardless of the objective probability of observing the event.

  26. Very recently, online sports and non-sports exchanges, initiated trading on contracts where account holders may buy or sell the future outcome of various events. Intrade.com, for example, allows to trade contracts that bet on the venue (region) that will host the Summer Olympics.

  27. If prices are affected by investor sentiment, regardless of the objective probability of observing the event, the behavioral effect is subsumed by parameters φ 0 and λ 0 .φ 0  + λ 0  > 0; φ 0  < 0; and—φ 0  > λ 0 /2 (asymmetrical effect).

  28. We looked upon other significance parametric and non parametric tests. Results are not shown in a table to save space. The significance of the results discussed in the paper is barely unchanged.

  29. There are some exceptions. For example, Greece experienced statistically positive abnormal of returns (+7.8%) regarding the announcement of the nomination to host the 2004 Olympic Games.

  30. Results for the other industries are not reported here to save space.

  31. The aggregate values shown at the bottom of Table 5 are equally-weighted averages of industry indices. The comparison of these equally-weighted values with the value-weighted averages reported in Table 4, show that market weights do not drive the results.

  32. This was a very tight victory: Germany secured 12 out of the 23 votes against the 12 received by South Africa (the other candidate in the final round).

  33. The minimum was 2.4% for the 2000 World Exhibition: Germany secured 21 out the 41 votes against the 20 received by Canada.

  34. We also run the regression with a balanced sample, including only those events for which we had information regarding the winning and the losing bidders (14 events, 28 observations -14 winners and 14 losers). The results are inconclusive.

References

Download references

Acknowledgments

Ana Paula Serra acknowledges the generous support of CEMPRE—Centro de Estudos Macroeconómicos e Previsão—a research unit financed by FCT—Fundação para a Ciência e a Tecnologia, Portugal through Programa Operacional Ciência, Tecnologia e Inovação (POCTI) of Quadro Comunitário de Apoio III, which is financed by FEDER and Portuguese funds.

Author information

Authors and Affiliations

  1. Faculdade de Economia da Universidade do Porto, Porto, Portugal

    António Miguel Martins

  2. CEMPRE, Faculdade de Economia da Universidade do Porto, Rua Dr. Roberto Frias, 4200-464, Porto, Portugal

    Ana Paula Serra

Authors
  1. António Miguel Martins
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ana Paula Serra
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Ana Paula Serra.

Appendix

Appendix

1.1 A model of market impact of partially anticipated events

This appendix presents a simple model of partial anticipation that generates a set of hypotheses tested in our study. The model is built upon the models proposed by Malatesta and Thompson (1985) and Edmans et al. (2007).

We assume that investors partially anticipate the likelihood that a particular country hosts one or several international sporting or cultural events. Let’s denote p i as the probability of country i hosting the event. Before the nominated host country is announced, the anticipated economic impact of the event is already reflected in the market valuations of the bidding countries. The economic impact of a particular event (its net present value) for candidate country listed firms is denoted by NPV. If country j is chosen to host the contest (the winning country), the (positive) market price effect at the date of announcement is given by (1−p j ) x NPV j . The observed market reaction is thus a biased estimate of the true economic effect and is inversely related to the prior probability of winning. As for the market impact of the candidate countries whose bids were rejected (the loser countries), we should observe a negative market effect following the announcement, and, in absolute terms, positively related with the prior probability of winning. If we focus on the two countries in the last round of the voting process, when the final outcome is announced, the market price effect for the losing country l is given by −p l NPV l that equals −(1−p j )NPV l .

The potential benefits brought by the organization event are expected to be different from country to country (NPV varies across bidding candidates) and consequently the absolute magnitude of the stock market effects to winners and losers can differ substantially. Therefore, a priori, asymmetric effects are expected for the winning and losing countries.

In assessing the probability of winning (losing), investors may consider the degree of competitiveness of the contest, whether the country is considered to be a front runner in advance and the initial rounds of the voting (that are publicized before the final outcome is realized). Our empirical model in Section 4 accommodates some of these features.

At the time of the announcement, two possible outcomes may result. Either the country wins the organization of the event or loses. Let’s denote VW as the market valuation of a particular country at time 0 if it hosts the event and VL its market valuation otherwise. Market valuation just before the event is announced (t = −1) for a candidate country is thus given by:

$$ V_{i - 1} = \frac{{E\left( {V_{i0} } \right)}}{{1 + E\left( {R_i } \right)}} = \frac{{p_i VW_i + \left( {1 - p_i } \right)VL_i }}{{1 + E\left( {R_i } \right)}} $$
(A-1)

Given that

$$ VW_i = VL_i + NPV_i $$

(A-1) can be rewritten as:

$$ V_{i - 1} = \frac{{VL_i + p_i NPV_i }}{{1 + E\left( {R_i } \right)}} $$
(A-2)

Abnormal returns at the date of the announcement can be computed as:

$$ AR_i = \frac{{1 + R_i }}{{1 + E\left( {R_i } \right)}} - 1 $$
(A-3)

where

$$ R_i = \frac{{V_{i0} }}{{V_{i - 1} }} - 1 $$

and E(R i ) is the company’s expected return for a given return generating process.

For the winning country j, V 0j = VW j , and abnormal returns are thus given by:

$$ \begin{array}{*{20}c} {AR_j = \frac{{{\raise0.7ex\hbox{${VW_j }$} \!\mathord{\left/ {\vphantom {{VW_j } {V_{j- 1} }}}\right.} \!\lower0.7ex\hbox{${V_{j- 1} }$}}}}{{1 + E\left( {R_j } \right)}} - 1} \\ { = \frac{{{\raise0.7ex\hbox{${VW_j }$} \!\mathord{\left/ {\vphantom {{VW_j } {{{\left[ {p_j VW_j + \left( {1 - p_j } \right)VL_j } \right]} \mathord{\left/ {\vphantom {{\left[ {p_j VW_j + \left( {1 - p_j } \right)VL_j } \right]} {\left( {1 + E\left( {R_j } \right)} \right)}}} \right. } {\left( {1 + E\left( {R_j } \right)} \right)}}}}}\right.} \!\lower0.7ex\hbox{${{{\left[ {p_j VW_j + \left( {1 - p_j } \right)VL_j } \right]} \mathord{\left/ {\vphantom {{\left[ {p_j VW_j + \left( {1 - p_j } \right)VL_j } \right]} {\left( {1 + E\left( {R_j } \right)} \right)}}} \right. } {\left( {1 + E\left( {R_j } \right)} \right)}}}$}}}}{{1 + E\left( {R_j } \right)}} - 1} \\ { = \frac{{VW_j }}{{p_j VW_j + \left( {1 - p_j } \right)VL_j }} - 1} \\ { = \frac{{VL_j + NPV_j }}{{VL_j + p_j NPV_j }} - 1} \end{array} $$

that can be rewritten as

$$ \begin{array}{*{20}c} {AR_j = \frac{{VL_j + p_j NPV_j + \left( {1 - p_j } \right)NPV_j }}{{VL_j + p_j NPV_j }} - 1} \\ { = 1 + \frac{{\left( {1 - p_j } \right)NPV_j }}{{VL_j + p_j NPV_j }} - 1} \\ { = \frac{{\left( {1 - p_j } \right)NPV_j }}{{V_{j- 1} \left( {1 + E\left( {R_j } \right)} \right)}}} \end{array} $$
(A-4)

For the losing country l, V 0l = VL l , and abnormal returns are given by:

$$ AR_l = \frac{{VL_l }}{{VL_l + p_l NPV_l }} - 1 $$

that can be rewritten as

$$ \begin{array}{*{20}c} { = \frac{{VL_l + p_l NPV_l - p_l NPV_l }}{{VL_ l + p_l NPV_l }} - 1} \\ { = 1 - \frac{{p_l NPV_l }}{{V_{l- 1} \left( {1 + E\left( {R_l } \right)} \right)}} - 1} \\ { = \frac{{ - p_l NPV_l }}{{V_{l- 1} \left( {1 + E\left( {R_l } \right)} \right)~}}} \\ { = \frac{{ - \left( {1 - p_j } \right)NPV_l }}{{V_{l - 1} \left( {1 + E\left( {R_l } \right)} \right)}}} \end{array} $$
(A-5)

Rights and permissions

Reprints and permissions

About this article

Cite this article

Martins, A.M., Serra, A.P. Market impact of international sporting and cultural events. J Econ Finan 35, 382–416 (2011). https://doi.org/10.1007/s12197-009-9087-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12197-009-9087-1

Keywords

Search

Navigation