Economic Aspects and The Summer Olympics: A Review of The Related Research
Economic Aspects and The Summer Olympics: A Review of The Related Research
Economic Aspects and The Summer Olympics: A Review of The Related Research
by
Evangelia Kasimati
University of Bath
United Kingdom
E-mail: E.Kasimati@bath.ac.uk
ACKNOWLEDGEMENTS
The author would like to gratefully acknowledge and thank Prof. John Hudson, Dr.
Peter Dawson, Adam George-Wood, Nikos Veraros, Martha McIntosh for their helpful
comments on earlier drafts of this paper. Special thanks also to two anonymous referees
who provided substantial and constructive comments. Finally, the author would like to
fund this research. Any remaining errors or omissions are the author’s alone.
Page 2
ABSTRACT
As the Summer Olympics are growing with larger media coverage and sponsorship,
host cities have started to attach great importance to the tourism and other likely
studies have been conducted to consider the various economic implications on the hosts.
This paper examines and evaluates methods and assumptions employed by the
economic studies. It also compares ex-ante models and forecasts with the ex-post
approach. The aim is to improve the information available to policy makers and
Page 3
INTRODUCTION
The modern Olympic Games were first held in Athens in 1896. Over the years, the
Games survived many trials including wars and boycotts, and each set of Games is held
every four years. In recent years, the interest of countries and regions in staging a future
edition of the Games has grown because of the perception that doing so would help
As well as the likely impacts on the socio-cultural and environmental areas, host cities
place great emphasis on the economic implications of the Olympics and the tourism
development. These implications have received increasing attention over the last two
decades, involving economic studies to provide a measure of the net gains which
hosting the Games may provide. While economic impact analyses prepared by or on
behalf of Olympic advocates have demonstrated economic advantages from hosting the
Games, potential host communities pose the question of whether, in fact, the economic
benefits of the Olympics are pragmatic and, if they are, the extent to which such
Much of the published literature of the Olympics emphasises long-term benefits such as
and Aitken, 1985; Hall, 1987; Kang, 1988; Robin, 1988; Walle, 1996; French and
Disher, 1997). In contrast, potential negative impacts include high construction costs of
public sports infrastructure and related necessary investments (usually putting a heavy
Page 4
burden on the government budget), temporary crowding problems, loss of visitors,
activities (Hiller, 1990; Darcy and Veal, 1994; Mount and Leroux, 1994; Leiper, 1997;
Spilling, 1998).
The objective here is to review the existing literature focused on the economic impacts
of the Summer Olympic Games. No economic impact studies were found for Games
before Los Angeles in 1984. Seven cases of the modern Games are examined, dated
between 1984 and 2012. Thirteen studies are considered that investigate various
economic variables related to the hosting of Games and they have been categorised into
This review does not attempt to draw any conclusion as to which Games have the most
favourable economic impacts. This would require a comprehensive study involving the
review, comparison and justification of the models from both theoretical and empirical
standpoints. Instead, our implicit objective is more modest. The goal is primarily to
provide an overview and evaluation of the different approaches and demonstrate the
The remainder of this article is organised as follows. It begins by explaining the link
between direct, indirect and induced economic effects, which is the principal theory
Page 5
Summer Olympics. Finally, the article analyses each study in turn, evaluates the
When a city is awarded the Summer Olympics, a large amount of new money is
expected to flow into the host economy and re-circulate throughout it. The economic
effect from hosting the Games occurs because an inflow of funds, which have not been
switched from elsewhere in the economy and probably would not otherwise have come
without the Games, will enter the local, regional or national economy. This inflow of
money stems from broadcasters, sponsors, Olympic family, athletes and dignitaries as
well as non-area travellers who would be defined as ‘tourists’ by those in the tourism
business.
There has been a tendency to assess the economic impact of Summer Olympics using
the ‘multiplier’ concept. Briefly, a multiplier estimates the number of times a unit of
currency, once spent within an economy, is re-spent within the borders of that economy.
The overall effect of the new money on the local/regional/national economy is broken
1. Direct Effect: the first round economic effect of the new money spent by outside
2. Indirect Effect: the subsequent rounds of the injected money within the
Page 6
3. Induced effect: the proportion of household income then re-spent in other
The indirect and induced effects together are collectively referred to as secondary
Figure 1 here
The multiplier analysis has been a common form of estimating the re-spending impact
in estimating the direct effect, those errors of calculation are compounded in estimating
essential in order for the economic impact estimates to be reliable (Baade and
Matheson, 2002).
The three most commonly reported multipliers are those of sales, income and
and secondary effect of the injected money on the business activity and turnover.
Household Income multipliers concentrate on the direct and secondary effects on the
household income. Employment multipliers measure the number of new full-time jobs
While the sales multiplier is the one most often used in the economic impact studies,
Crompton (1995) argues that the household income multiplier is the most relevant for
assessing the economic impact of hosting a sport event. The reason for this is because it
Page 7
focuses particularly on the effect of the injected money on residents’ income and their
standard of living. In other words, the host community is not interested in knowing how
many sales are attributable to the hosting of the Summer Olympics, but rather what
On the other hand, the employment multipliers are the least reliable among the others
(Fletcher and Snee, 1989 cited Crompton 1995, p.22). Their basic assumption of full
utilisation of existing employees may creates errors in calculating the increase in the
level of employment, particularly for ‘one-time’ mega sports events like the Summer
Olympic Games. The short duration of the Games does not necessarily justify the hiring
of new employees, the generation of permanent full-time jobs and the sustainability of
the employment effects. Entrepreneurs will probably exhaust other alternatives like
asking existing employees to work overtime or perform other tasks, before hiring
additional work force to satisfy the temporary high demand (Crompton, 1995).
A short review in the literature reveals that the multiplier is a particularly contentious
measure. A study by Hunter (1988 cited Crompton 1995, p.33) argues that “economic
impact studies based on multipliers are quite clearly an improper tool for legislative
with integrity”.
In event economic studies, problems usually arise when researchers do not clearly
identify what type of multiplier (sales or income) is used in their methodology, and as a
Page 8
result misleading conclusions can be derived from the data. Because sales multipliers
include higher numbers compared to income multipliers, they tend to be attractive tools
for sport events’ advocates to use, in their attempt to justify the economic benefits of
hosting the events (Crompton, 1995). In addition, misapplication of the data may arise,
when spending generated by local residents or which occurred outside is included in the
overall economic effect. Furthermore, spending, either from tourists who re-scheduled
their already organised trip to coincide with the Games or from those who visit the host
for other reasons but also end up attending, is crucial to be excluded from an economic
In order for economists to identify and quantify the economic consequences of hosting
an event, such as the Summer Olympic Games, a modelling approach must be adopted.
In the published literature examined, two main approaches have been used under the
broad label of the Input-Output (I-O) and the Computable General Equilibrium (CGE)
framework.
The I-O method is a long-established technique originated by Leontief in the 1940s and
since then it has been very widely applied in economics. Classic I-O models are
structured around the input-output tables and its production or price identities, but make
I-O models is limited by the disaggregation of the published input-output table. Since
these models account for intermediate exchanges, they are useful for assessing industry
level impacts for changes in final demand, indirect tax rates or commodity price shocks
Page 9
(West, 1995). However, projections are normally made by specifying final demands
consumption, prices and income are determined with strict identities. Consequently,
there is no integration between final demand and prices or income and no guarantee that
there will be economic consistency among, for example, consumption, prices and
output often leads to severe instability problems (Almon, 1966; Steenge, 1990).
Studies that adopted the I-O analysis to evaluate the total economic impact of hosting a
mega sporting event made use of linear assumptions. They calculated a set of
multipliers suggesting particular proportions of consuming the inputs and used them
intact, no matter of the scale of the injected funds and the surge in the economic
activity. As a result, they failed to take into account economies of scale, production
close to full capacity and price adaptations to demand changes. Ignoring these factors
System (RIMS II), a computer program often used by studies examined the Summer
Olympics in the USA. RIMS II has been proven to be successful for measuring effects
at several levels of industrial aggregation, when initial tourist spending is known, but
fails to examine the effect on nearby areas, since it is a single-region model (Humphreys
and Plummer, 1995). An alternative I-O computer program, also developed in the USA,
Page 10
Although a comparatively large number of the referenced economic studies have been
carried out in an I-O framework (see Table 1), studies around the Sydney Olympics
turned towards the use of CGE models. CGE are disaggregated representations of the
economy, which use input-output structure for the production side of the economy.
structure and behavioural functions. Normally, however, behavioural parameters are not
estimated with regression analysis but are deduced from the single year’s set of data or
assume flexible prices that move to clear all the markets simultaneously (although some
CGE models will assume some sticky prices, such as in the labour market).
Earlier CGE models were used to estimate different static equilibriums under Walrasian
general equilibrium theory. Most contemporary CGE models have been expanded to
model, used by Australian studies to measure the economic impact of the Sydney
The MMRF used the so-called ‘bottoms-up’ approach. A number of regional economic
models are included and then are linked using interregional flows of commodities,
factors of production and population. The bottoms-up approach allows the modelling of
economic agents’ behaviour at the regional level and then their aggregation is
Page 11
states and territories and generates results for all regions in a steady multiregional
accounting framework, its size limitation hinders the application of a similar model to
Due to vague technical details often found in the economic studies, a deep penetration
proved to be a difficult task. The economic models rely on assumptions that reduce the
its own limitations defined by its assumptions. Most of the theoretical assumptions
employed in MMRF, such as perfect competition in product markets, zero pure profits
and constant returns to scale production functions, labour market equilibrium, are not
always valid for the Australian states. It is therefore important to consider whether these
In the case of I-O analysis the assumption that the I-O coefficients remain unchanged or
can be extrapolated into the future in a reliable manner is of particular importance. This
is still more so, when the I-O model is being used to analyse the impact of major
structural changes or shocks such as that of hosting a mega sporting event. The import
perhaps to all forms of analysis, is differentiating between the short and the long-term
impact of hosting the Games. For example, the examination of the extent to which the
employment generated is sustainable into the long run. From the short overview,
however, it is our understanding that I-O model has been comparatively more popular,
Page 12
EX-ANTE AND EX-POST ECONOMIC IMPACT ASSESSMENTS
The importance of the relationship between tourism and the Summer Olympics has
gained increased recognition in recent years. The tourism effect is one among several
that bid and host cities seek, arguing that the international media coverage preceding
and during the Games presents a tremendous opportunity to advertise themselves in the
global marketplace.
In an attempt to assess the likely growth in tourism as well as other economic effects,
ex-ante assessments have been carried out to forecast the impacts of the Summer
Olympics. Table 1 shows that a number of ex-ante economic analyses have been
conducted, while the research significantly lacks ex-post impact assessments. An ex-
post analysis examines the economic situation of the geographical influence zone before
and after the event and manages to isolate the event from other factors that may run at
the same time and may contributed to the economic impact (Baade and Matheson,
2002).
Olympic process, and the reader must bear in mind that the report writers were
potentially motivated to come up with a favourable result (Baade and Matheson, 2002).
This interpretation could more likely be the case when reports were prepared to justify
an Olympic bid. A good way to provide some balance to these views would be to read
economic impact studies prepared by ‘anti-Olympics’ groups, but there are none
currently available. There is, however, a growing non-affiliated literature that can be
used as a counterweight.
Page 13
For example, the anti-Olympic alliances ‘Bread Not Circuses’ (BNC) and ‘People
Ingeniously Subverting the Sydney Olympic Farce’ (PISSOFF), based on Toronto and
Sydney respectively, made use of the Internet to promote their Olympic critique (current
argument was that the public money spent for the Games would be taken from other
more important sectors (e.g. education, health, environment, prosperity). Now with
almost every potential Olympic city’s bid there tends to be the creation of an anti-
Olympic alliance such as the recent example from Vancouver's bid for the 2010 Winter
group and aims to ensure that the environmental, social, economic and civil rights issues
On the other side, the bidding process itself has gained attention. In his book, Hill
Manchester to host the 1992 and 2000 Olympic Games respectively, focusing especially
on the politics involved. Whilst Hiller (1999) has discussed the strategy employed by
Cape Town in its bid to host the 2004 Olympic Games. Further dimensions of the
bidding process, however, have been revealed by assertions of bribery and corruption.
Books such as ‘Lords of the Rings’ (Simson and Jennings, 1992) and ‘New Lords of the
Rings’ (Jennings, 1996) criticised intensively the legitimacy of the bidding process,
claiming that IOC members corruptly requested bribes and accepted generous gifts from
potential host cities in return for their votes. In addition, the Salt Lake scandal further
Page 14
emphasised the need to address such problems (McIntosh, 2000) and virtually prompted
a revamping of the IOC’s rules with respect to the host city bidding process.
The review will now analyse the studies mentioned in Table 1 with reference to a
specific question: What are the economic implications of the Summer Games on the
host?
Three studies commissioned on the Sydney Olympics predicted the event would
generate substantial extra revenue for Australia, and New South Wales (NSW) in
Table 2 here
Although KPMG (1993) adopted a different modelling approach, its figures broadly
concurred with those released by Andersen (1999) and NSW Treasury (1997). The I-O
framework used in KPMG’s study ignored supply-side constraints and therefore made
crowding out, price increases due to resource scarcity and public financing of
Olympics and the consulting firm should take this into consideration.
Investigating the tourism impacts of the Games, Andersen and NSW Treasury gave
little consideration to the likely loss of visitors as a result of hosting the 2000 Games.
This subject is of particular interest in the light of the argument put forward by Leiper
Page 15
(1997), which mentions that while mega-events such as the Summer Olympics may
encourage new tourists, the holiday-makers, business travellers or even local residents
A number of ex-ante studies are also available for the next Summer Olympics, which
will be taking place in 2004 in Athens, Greece; prominent amongst these are the studies
by Balfousia et al. (2001) and Papanikos (1999). Balfousia et al. had the advantage of
utilising the most recent estimates of the direct impacts of the Games, including updated
estimates for the Olympics budget. However, scepticism is raised regarding data
Table 3 here
Despite major methodological differences between the two studies, their results do not
differ significantly with both suggesting growth in tourism and revenue. The
scenarios in macroeconomic settings, but failed to take into account possible resource
constraints. On the other hand, Papanikos ‘borrowed’ the value of multipliers from
other studies in related cities. This probably happened because the direct estimation of
the value could be both complicated and costly. However, it might affect his results,
since economic relationships may be different between communities. Both studies make
predictions on a national level and lack an explicit spatial dimension in assessing the
impact of the Games. The choice of the nation as a reference area is doubtful, because
Page 16
as Howard and Crompton (1995) illustrated, the larger the assessed area, the smaller the
leakages that are likely to happen and then the larger the multiplier is likely to be.
evaluation of the Balfousia et al. model as there are few published details relating to its
theoretical structure.
Other ex-ante studies that have looked at the economic impact of Summer Olympic
Games are those of Brunet (1993, 1995) and Kim et al. (1989). Quantifiable data
describing expenditures, contracts, jobs, investments and tourism were based almost
exclusively on secondary research and the studies did not provide any predictions using
a type of modelling. Rather they were more theoretical in their approach aiming to
identify and collate evidence of the economic benefits of the Barcelona and South
Korea Games respectively. It is our understanding that the studies were conducted with
An attempt to offer an ex-post economic impact analysis of the Summer Olympics was
made by Baade and Matheson (2002). Their aim was to assess changes in employment
in Los Angeles and Atlanta that were attributable to the staging of the 1984 and 1996
Olympics respectively. In other words, their ex-post approach was targeted to estimate
the level of employment in the Games’ absence. To achieve this, they constructed an
capita personal income, wages, taxes as well as dummies variables for the occurrence of
the Olympics and the oil boom. Using standard regression analysis techniques, Baade
Page 17
and Matheson found that the coefficient for the Olympics variable was insignificant.
The econometric equation was then used to estimate changes in employment and
isolated the contribution of the Games by comparing this estimated value with the actual
employment levels.
Their results for employment growth were by far divergent with those released by ex-
ante studies of Economic Research Associates (1984) and Humphreys and Plummer
(1995) and have brought to light possible overestimation reported by the latter studies.
Another key finding was that the economy virtually returned to its ‘normal’ pattern
afterwards and any increase in economic activity attributable to the Games was
temporary. However, Baade and Matheson’s conclusions rest heavily on the model
being correctly specified, which invites one to wonder how sensitive these results are to
alternative specifications.
Covering the period of 1984 through 2012, all the ex-ante economic studies indicate the
significant role of the Summer Olympic Games in the promotion of the host economy.
They highlighted the extension of the Games economic impact well beyond the actual
period of the event occurrence itself. Economic growth, increased tourism, and
potentially biased, because the ambition of those commissioning the studies is to favour
the hosting of the Games. This issue has received a great deal of attention from scholars
investigating the Games and other mega-events (Mills, 1993; Crompton, 1995; Howard
Page 18
and Crompton, 1995; Kesenne, 1999; Porter, 1999; Preuss, 2000; Baade and Matheson,
2002). Nevertheless, it is our opinion that if the estimation process is made transparent,
then the findings are reliable. Taking into account the strengths and weaknesses of all
the methods and techniques used, the discussion here shows that ex-ante models and
forecasts were not confirmed by ex-post analyses and this therefore prompts the need
Research in this field needs to further consider a substantial element, which is the
opportunity cost involved in hosting the Summer Olympic Games or other mega-events.
Host communities often pose the question of whether financing the Games is the most
effective and efficient use of the public money. In other words, if the public funds spent
on the Games were used in a different way, would the host economy receive a greater
return than it does when these funds are spent on Games investments? To answer this
one need look no further than Kesenne’s argument (1999) that even though a mega-
event does create net benefits, public funding should occur only if the mega-event yields
unfeasible to measure the net benefits of all possible alternative projects; however some
economic retreat after the completion of the Games. When the level of income and
investments falls after the event, then the multiplier also follows. To illustrate this point,
findings from broader mega-event literature could be utilised to demonstrate that ‘one-
Page 19
time’ events have no lasting post-event effects in new business activities or employment
shortcomings found in earlier ex-ante and ex-post studies and that they concentrate on
areas which most need the effort. This will help planners and potential hosts of mega-
Page 20
Evangelia KASIMATI, Figure 1
New Money
Injected into
Accommodation
Food Transport Games tickets
a. b. c.
d. DIRECT
Inter- industry purchases Direct household Government
Leakages EFFECT
within the economy income revenue
ALL BUSINESSES
a. b. c.
d. INDIRECT
Inter- industry purchases Secondary household Government
Leakages EFFECT
within the economy income revenue
chosen to show how the multiplier concept operates, but should be similarly
implemented for ‘food’, ‘transport’ and ‘Games tickets’. The three direct
spend this money in the same four ways, generating the indirect effect.
Leakages occurred since some money could be spent outside the host
economy. Moreover, some of the household income could leak out of the
Page 21
Evangelia KASIMATI, Table 1
Page 22
Evangelia KASIMATI, Table 2
Table 2: Sydney Games Impact Summaries. Sources: KPMG, 1993; NSW Treasury/CREA,
1997; Andersen/CREA, 1999
Projected Figures Andersen/CREA NSW Treasury/CREA KPMG
Sponsor of Analysis Sydney OOC NSW Treasury Sydney Bid
Committee
Addition to Australian GDP A$96 6.5 billion A$96 6.4 billion A$92 7.3 billion
Addition to NSW’s GDP A$96 5.1 billion A$96 6.3 billion A$92 4.6 billion
International Arrivals in Australia 1.5 million 2.3 million 1.3 million
Additional Tourist spending A$96 2.7 billion A$96 4.3 billion A$92 3.0 billion
New Jobs 90,000 98,700 156,198
Australia Australia Australia
Period 1994-2006 1994-2006 1991-2004
Page 23
Evangelia KASIMATI, Table 3
Page 24
REFERENCES
Airola J, Craig S. 2000. The Projected Economic Impact on Houston of Hosting The
Almon C. 1966. The American Economy to 1975. Harper and Row: New York.
Arthur Andersen. 1999. Economic Impact Study of the Sydney 2000 Olympic Games.
Baade RA, Matheson V. 2002. Bidding for the Olympics: fool’s gold? In Transatlantic
Sport: the comparative economics of North America and European Sports, Barros
of the Athens Olympic Games (in Greek). Centre of Planning and Economic
Research: Athens.
Resources, Financing, And Impact. In The Keys To Success, Miquel MD, Botella M
Crompton JL. 1995. Economic Impact Analysis of Sports Facilities and Events: Eleven
Darcy S, Veal AJ. 1994. The Sydney 2000 Olympic Games: The Story So Far.
1984 Olympic Games In Los Angeles And Southern California. Los Angeles
Page 25
French SP, Disher ME. 1997. Atlanta and the Olympics: A one-Year Retrospective.
Fuller SS, Clinch R. 2000. The Economic And Fiscal Impacts of Hosting The 2012
Hall CM. 1987. The Effects of Hallmark Events on Cities. Journal of Travel Research
26(2): 44-45.
Haxton PA. 1999. The Perceived Role of Community Involvement in the Mega-Event
Hosting Process: A case study of the Atlanta 1996 and Sydney 2000 Olympic Games.
Hill CR. 1996. Olympic Politics. Athens to Atlanta 1896-1996. Manchester University
Press: Manchester.
Hiller HH. 1990. The Urban Transformation of a Landmark Event: The 1988 Calgary
Hiller HH. 1999. Mega-events and urban social transformation: Human development
and the 2004 Cape Town Olympic bid. In The Impact of Mega-events, Andersson
Howard DR, Crompton JL. 1995. Financing Sport. Fitness Information Technology:
Morgantown.
Humphreys JM, Plummer MK. 1995. The Economic Impact on The State of Georgia of
Hosting The 1996 Olympic Games. Selig Center for Economic Growth: Georgia.
Jennings A. 1996. The New Lords of the Rings: Olympic Corruption and How to Buy
Page 26
Kang HB. 1988. Accelerating the Future-State: Urban Impact of Hosting the 1988 Seoul
Olympic Games. In: Hosting the Olympics: The Long-Term Impact, Conference
Report. East Asian Architecture and Planning Program, MIT and Grad. School of
Kim JG, Rhee SW, Yu JC, et al. 1989. Impact of The Seoul Olympic Games on National
KPMG Peat Marwick. 1993. Sydney Olympics 2000: Economic Impact Study. Sydney
Leiper N. 1997. A Town like Elis? The Olympics: Impact on tourism in Sydney. In
McIntosh MJ. 2000. The Olympic Host City Bid Process: Facing Challenges and
Mills ES. 1993. The Misuse of Regional Economic Models. Cato Journal 13(1): 29-39.
Study of the Impact of the Olympic Games on the Calgary Business Sector.
New South Wales Treasury. 1997. Research and Information Paper: The Economic
Tasmania: Australia.
Page 27
Papanikos GT. 1999. Tourism Impact of The 2004 Olympic Games (in Greek). Tourism
Preuss H. 2000. Economics of The Olympic Games: Hosting The Games 1972-2000.
Ritchie JRB, Aitken EC. 1984. Assessing the Impacts of the 1988 Olympic Winter
Games: The Research Program and Initial Results. Journal of Travel Research 22(3):
17-25.
Robin D. 1988. Hosting the Olympic Games: Long-Term Benefits to Sport and Culture.
In: Hosting the Olympics: The Long-Term Impact, Conference Report. East Asian
Simson V, Jennings A. 1992. The Lords of the Rings: Power, Money and Drugs in the
Spilling OR. 1998. Beyond Intermezzo? On the long-term Industrial Impacts of mega-
events: The case of Lillehammer 1994. Festival Management and Event Tourism
5(3): 101-122.
Walle AH. 1996. Festivals and Mega-Events: Varying Roles and Responsibilities.
Page 28
Werling JF. 1992. MIDE: A Macroeconomic Multisectoral Model of the Spanish
Page 29