Predicting the costs…

P R E D I C T I N G  T H E  C O S T S
M E G A – S P O R T I N G  E V E N T S :
M I S J U D G E M E N T  O F  O LY M P I C
P R O P O R T I O N S ?

Jonathan Barclay

The economic benefits of hosting mega-sporting events are often exaggerated. Ex-ante impact studies typically overestimate the gains and underestimate the costs involved. It is therefore difficult to explain in economic terms the intense competition among cities to hold such events.

Keywords: Cost–benefit analysis, impact studies, London 2012, multipliers,
Olympic Games, urban regeneration, World Cup,


In recent years cities have competed
vigorously for the right to host what can be
labelled as ‘mega-events’, namely the
quadrennial Olympic Games and FIFA
Football World Cup. One can afford such a
description when one considers the scale of
these sporting extravaganzas, with in-person
attendance in the millions and television
audiences in the billions.1
There are a variety of reasons why cities
may wish to host these events, the most
compelling being the promise of a vast
economic windfall forecasted by economic
impact studies. Given these forecasts, an
increasing number of developing economies
have joined the bidding frenzy, insisting on
their right to receive a share of the monetary
spoils and hopefully kick-start their
development. It is also evident that cities that
host these events must commit a significant
investment into sports stadia and other
miscellaneous infrastructure. Therefore, the
question is whether the economic benefit
compensates for and outweighs the vast costs
and substantial risks incurred. Are the games
‘fool’s gold’ (Baade and Matheson, 2002) or a
lottery jackpot (Preuss, 2006, p. 183)?

Olympic-sized costs

The Olympic Games are becoming an
increasingly expensive affair, especially for
host cities that see it not only as an
opportunity to construct new sports stadia,
but also to improve other infrastructure such
as communication systems, housing facilities
and traffic networks. Given these extensive
costs, it is not surprising that the majority of
host cities for the summer Olympics have
come from developed nations. Out of the 25
summer Olympics between 1896 and 2004, 14
were held in Western Europe and seven in the
United States, Canada and Australia
combined. The Games hosted by Mexico City
and Seoul (1968 and 1988, respectively) were
the only ones held in developing nations.
Indeed, in addition to investing large
sums in the construction of sports arenas,
Barcelona (1992) and Seoul (1988) used the
Games to upgrade their entire urban
infrastructure (ibid.). This displays organisers’
maxim of creating long-run benefits that
subsume short-run costs. Even Atlanta (1996),
which built relatively few sports arenas to
complement existing infrastructure for its
Olympic Games, cost an estimated $600
million (Baade and Matheson, 2004).
The FIFA World Cup presents a similar
situation. FIFA requires that a host nation
have at least eight, but preferably ten, modern
stadia with seating capacities in the range of
40,000 to 60,000. For the 2002 World Cup,
South Korea spent in the region of $2 billion
constructing ten new stadia and Japan
$4 billion in building seven new stadia and
refurbishing three existing ones. Moreover, it
is clear that other costs are also growing, with Salt Lake City,
host of the 2002 Winter Olympics, spending more than $300
million on security alone, and Greece, host of the 2004
Summer Games, more than $1 billion (Baade, 2006, p. 177).
This is in addition to the $1 billion estimated operating cost of
running a mega-event (Baade and Matheson, 2004).

Sports ‘boosters’ predictions

Irrespective of these costs, sports ‘boosters’ have predicted
large economic windfalls for cities hosting these ‘mega-events’,
envisioning multitudes of sports fans frequenting the city’s
restaurants, hotels and other businesses, spending vast
amounts of money. Olympic boosters have shown great
optimism. The Atlanta Olympic Organizing Committee
predicted a $5.1 billion boost and an increase in employment
of 77,000. The Sydney Olympic Games was predicted to have a
$6.3 billion impact as well as create 100,000 new jobs. For the
World Cup in the USA in 1994 Goodman and Stern predicted a
boost of $4 billion to the US economy, and in 2002, Dentsu
Institute for Human Studies estimated a $24.8 billion windfall
for Japan and $8.9 billion for South Korea, amounting to 0.2%
and 2.2% of their respective GDPs (Baade and Matheson,
2002). The academic premise behind these reports is that
these expenditures on mega-event infrastructure should be
considered investments that trigger positive economic returns
(ibid.). This has been sharply criticised by many scholars,
firstly on the basis that many of these studies are
commissioned by those who have a vested interest in holding
such events (for example, standing to benefit directly from the
provision of public subsidies that these reports may influence
or justify).

Methodological errors associated with ex ante studies

It should be noted that these ‘impact’ studies (which provide
the rationale for funding) are themselves predictive or ex ante.
They derive the economic ‘impact’ in two main ways: from the
effect of the construction of infrastructure such as sporting
facilities, and from the total commercial activity that takes
place during the event, which takes into account an estimate of
the number of visitors, the number of days a visitor is
expected to spend and how much on average he or she will
spend (Matheson, 2006a). In the latter case, these figures are
combined to estimate a ‘direct economic impact’. However, the
method of estimating this ‘direct impact’ has been rejected by
many scholars as fundamentally flawed. One explanation for
this is that the attributable increase in direct spending may be
deduced using a ‘gross’ measure instead of a ‘net’ one. Since
most consumers have relatively inflexible leisure budgets,
spending on an event such as purchasing tickets substitutes for
other expenditures on other activities in the local economy
such as theatres, amusement parks or concert halls. This
provides one of the main reasons why impact studies are so
grossly exaggerated (Owen, 2005). The local consumer’s
expenditure is not a new economic activity but a reallocation
of spending that would have occurred if the event were not
held. Hence, in net terms, the effect on the local community is
likely to be zero. This effect has been demonstrated
empirically in relation to local sports teams and stadia (Coates
and Humphreys, 1999). To make studies more reliable for
mega-events such as the Olympics one might suggest surveys
carried out on those attending the event with questions
relating to place of residence thus enabling analysts to
eliminate those who hail from the local area (Baade and
Matheson, 2004).

Problems with analysing external visitors

But the inclusion of only external visitors may also result in
exaggerated figures if those attending an event at the Games or
the World Cup are also in the host city for alternative reasons
other than the event itself. These outsiders may be hosted by
their families or put up by their own companies, meaning that
spending on a sports event may just represent a reallocation of
leisure spending. Had the visitor not gone to the event, money
would have been spent on other attractions in the city in the
event’s absence. For example, if a man went to Beijing on
business for a few days and on one night went to watch a
sporting event, his net expenditure would be the cost of the
ticket only, as he would have made his other expenditures
(such as staying at a hotel and frequenting a restaurant)
regardless of his attendance of the event. It is clear that impact
studies may report the direct net economic impact to be far
greater given the businessman’s spending on hotels, meals
and such like (Siegfried and Zimbalist, 2002). In addition,
another phenomenon not adjusted for is the engaging of
‘time-switching’, which occurs when a visitor wishes to visit
the city in question, but arranges the trip to coincide with the
sporting event. By virtue of the fact that the event did not
influence the visitor’s choice of location, one cannot attribute
a net increase in spending as a result of the sporting event
(Crompton, 1995). However, it is evident that for large events
such as the Olympic Games, a high proportion of visitors are
likely to come from overseas on a special trip, resulting in a
greater convergence between net and gross spending.

Problems with accounting for those locals
who are non-attendees

Although amendments to the problems listed above may be
useful for those running the events, these impact studies will
continue to neglect the effect of these events on those residents
who do not attend but live in their vicinity. Indeed, many
residents may dramatically alter their spending patterns to
avoid either the inflated prices charged during the event or
congestion caused by its visitors. Baade and Matheson (2004)
thus state that a significant problem with economic impact
studies is not information relating to direct expenditures but
the lack of it with regard to the pattern of economic activity of
those who do not attend the event.

‘Crowding out’

Moreover, as a large proportion of mega events such as the
Olympics are held in popular tourist areas, the negative
externalities caused by an event, such as congestion, may
dissuade regular non-interested tourists from visiting the city
during the event. If local restaurants and hotels are near full
capacity, sporting visitors may actually displace and ‘crowd
out’ regular tourists, resulting in a smaller than predicted net
impact. One prime example is the effect of the FIFA World
Cup on tourism in South Korea, where it was estimated that
the total number of foreign visitors during that time was
identical to that in the same period in the previous year. Also
in terms of consumer expenditure, USA Today in 2002 did
report that although ‘Consumer goods such as sporting goods
sold well . . . some casinos and hotels had drop-offs as regular
customers and travellers avoided World Cup hassles’ (Baade
and Matheson, 2004). From data and anecdotal evidence, it
was evident that the Olympics had significant crowding out
effects on Atlanta. According to French and Disher (1997), in
parts of town not near the Olympic Park ‘many hotels and
restaurants reported significantly lower than normal sales
volume during the Games. Even shops and resorts in areas up
to 150 miles away reported slower than normal business in the
summer of 1996’.

Misuse of multipliers

One of the more difficult problems for economic forecasters
lies in the misuse of multipliers. ‘Direct’ expenditures
estimated to be a result of the event are used to deduce
‘indirect’ effects, which are more often than not prone to
exaggeration. The economic multipliers that are generally used
by economic forecasters are based on complex formulae which
model the relationships between industries within the region.
But Matheson has suggested that during a mega-sporting
event such relationships are unlikely to hold, thus rendering
the multiplier inaccurate. There is also great difficulty
accounting for the various leakages that might occur.
Evidently, one has to differentiate between whether visitors
spend their money within the local economy or on hotel
rooms and restaurants, which themselves belong to national
chains (Matheson, 2006b). In the latter case, profits earned
during an event do not flow to the local economy but to
shareholders throughout the world.

Supply-side leakages

These leakages are further increased by the temporary entry of
external firms selling products during the Games. Indeed,
these may mean local residents do not actually benefit from
the supposed growth in tourism during the event. To some
extent the Olympics are self-contained, as many sponsors and
corporations are allowed to have access to prime venues within
the Olympic Park which local businesses do not (Owen, 2005).
This makes the Olympics an economy unto itself, meaning
that much income would flow to firms that are not permanent
elements of the local economy. Substantial leakages also
accrue from those involved in staging the Games and
businesses that provide goods and services during the event.
One example is that if the local economy is running at close to
full employment, hotels and restaurants that require extra
labour to deal with greater than normal levels of tourists will
hire from external communities where there is a surplus.
Evidently, a substantially lower proportion of the wages that
are paid out in these cases will be recirculated in the local
economy. Such theoretical problems arise because, instead of a
balance of payments method, forecasters tend to use
input–output models, such as the US Department of
Commerce’s Regional Input–Output System (RIMS II), which
do not account for subtleties such as full employment (Baade
and Matheson, 2002, p. 11).

Construction: more a cost than a benefit?

The example above is also true in the construction process for
the Games, which itself provides another dubious area in
economic impact studies (for example, Siegfried and Zimbalist
noted that workers are more likely to live outside the local
area). Moreover, the writings of many academics (Coates and
Humphreys, 1999; Noll and Zimbalist, 1997) have found no
correlation between sports stadia construction and economic
development. Yet many studies consider the construction of
stadia a benefit as opposed to a cost. Although new
construction may increase economic activity, it is also
necessary to consider the vast opportunity costs, as public
expenditure on such projects would mean a reduction in other
public services, greater government borrowing or higher levels
of taxation. Would the return on a sports stadium exceed that
of an alternative use of resources?
If the host city has high unemployment, the building of
stadia might be considered an efficient way to get idle people
into work. However, this cannot be a net benefit given the fact
that it represents only a transfer of workers from one job to
another. Noll and Zimbalist describe this as just as beneficial
as if the government ‘employ half the workers to dig a hole
and the other half to fill it up’. One can also consider
construction employment as temporary, and although many
in South Africa expect employment to rise and attract
migrants from rural areas for the 2010 World Cup, urban
unemployment is expected to rise after the tournament (Pillay
and Bass, 2008). Thus, according to Hiller, mega-events can
become a bastion of ‘job creation . . . even if it . . . [is]
unproven and perhaps wishful thinking’. In an ex post study of
the Atlanta Games of 1996, Baade and Matheson found that
employment only increased between by 3,500 and 42,000,
only a fraction of the organising committee’s estimate of
77,000 new jobs. For the 1984 Games in Los Angeles, the
increase in employment, found to be 5,043, was concluded to
be transitory, with the same researchers’ model failing to find
any net employment gains as a result of the Games (Baade and
Matheson, 2002, p. 31). However, this result may be due to the
fact that Los Angeles made no significant investment in
sporting infrastructure for the Games, maximising use of
existing facilities.

Legacy and white elephants

Such employment outcomes suggest dubious legacy values of
both recent US Games. The extent to which the facilities are
used in the future is extremely important, especially the newly
constructed sports stadia. Evidently, sports are essentially a
luxury good, and thus demand is likely to decrease
substantially after a mega-event. There is therefore a great risk,
especially for developing countries, that infrastructure for
mega-events can become white elephants. For the 2003
All Africa Games, the Nigerian government spent around
$300 million on a new 60,000-seater stadium amongst other
venues, yet the country remains blighted by shortages of fuel,
frequent blackouts, bad roads and a high crime rate.
Moreover, there is a limited amount that one can do with an
empty football stadium. After the 2002 World Cup, only five
out of the ten new stadia in South Korea had regular tenants.
In addition, professional football attendances there average
only 3,000, a tiny fraction of a 40,000–60,000-seater stadium.
Even in Japan, with a more developed football league, J-League
crowds average only 16,000. Baade and Matheson, from their
own empirical evidence, claim that the economic impact of
such events is thus ‘transitory’ and ‘one-time’ rather than ‘a
steady state change’. Hence it seems that the only way that an
event can have a positive lasting effect is if its infrastructure is
able to exist symbiotically with that in the surrounding
economy, neither competing for nor displacing existing capital
and labour.

Hidden costs

In addition, countries are faced with other hidden costs that
may not have been accounted for in impact studies. These may
include the continual upkeep and maintenance costs of large
stadia. Whitson and Horne (2006) have concluded that social
goals from the 2002 World Cup in Japan ‘that might follow
from high-quality public infrastructure for sport . . . are still
far from being accomplished’.
Other sectors, such as the hotel industry, are also likely to
be hit by long-term costs. In order to accommodate the
multitude of visitors during an event, hotels may have a
construction boom to increase their inventories. However,
when the event is over, demand for high-end accommodation
will fall, leaving too-much surplus capacity and putting
downward pressure on room-rates (Humphreys and
Prokopowicz, 2007). This could have disastrous effects such as
those displayed in Lillehammer after the 1994 Winter
Olympics, where it was reported that within five years of the
event, 40% of full-service hotels in the region had gone
bankrupt (Teigland, 1999).

Housing and urban regeneration

Although there may be significant long-term benefits from
schemes to improve city infrastructure (Barcelona created new
districts, new trunk roads, renovated old districts and reduced
overcrowding in the city (Barget and Gouguet, 2007, p. 168)),2
these may have limited effects on less affluent people. In terms
of housing, as flagged up by Pillay and Bass in their
assessment of South Africa’s 2010 FIFA World Cup
development proposals, the positive effect on property prices
in an Olympic area will have a negative effect on the poor who
live there; rents may increase until they become unaffordable.
Although the area benefits from the investment in terms of a
growing property market and the improved infrastructure
associated with a stadium (transport links, for example), the
people it was intended to benefit might be pushed elsewhere.
Moreover, this may mean that the intended regeneration of the
host city may amount only to a redistribution of people, as
those who originally resided in the Olympic area move
elsewhere, bringing poverty and social problems with them.
This redistribution will therefore be zero-sum or actually
negative-sum, once one factors in the cost to the taxpayer in
terms of public subsidies for the new housing or transport
links. Indeed, there is strong evidence that this process has
operated in the lead up to several mega-events, often as a
result of direct state coercion. It has been estimated that
700,000 people were forcibly evicted in the lead up to the 1988
Seoul Olympics, and a staggering 1,500,000 for the Beijing
Olympics.3 Indeed, the Centre on Housing Rights and
Evictions has estimated that in the past 20 years, the Olympics
have displaced over 2,000,000 people.4

Intangible factors?

If the economic benefits for hosting mega-events are doubtful,
there must be other intangible reasons that explain why
countries wish to host them, even with the difficulties they
pose. One line of argument is the perceived status benefit
where a city can rise in the hierarchy of ‘world cities’, making a
claim to high global standing. This seems appealing amidst
the growing perception of competition for footloose capital
flows and tourism in the current phase of globalisation. It is
clear that in hosting the 2008 Olympics, Beijing hoped to join
the top tier of cities in the world and surpass its Asian
competitors: Tokyo, Singapore and Hong Kong. Moreover,
benefits are described in such ways as ‘a restoration of
self-confidence’, ‘civic pride’ and ‘dynamism’. Therefore, major
sporting events become part of what was labelled as the ‘Big
Kahuna approach’, that, with the hope of large-scale
construction, ‘cash and prestige’ will be brought to the area
(Greene, 2003, p. 165). Other intangible ‘benefits’ include
objectives such as ‘nation building’, for example the lasting
image of Nelson Mandela presenting the Rugby World Cup
Trophy to François Pienaar in 1995, an event held in South
Africa. International politics play an extremely important role
as well. To the Chinese, the Olympics in 2008 were less of a
sporting spectacle and more of a ‘coming out’ party, which
showcased the economic and political development of China.
Indeed, Wen Jiabao stated in April 2008 that the Beijing
Olympics present an opportunity for China to show the world
how ‘democratic, open, civilised, friendly, and harmonious’ it
is.5 However, cynics may make comparisons with the ‘political
theatre’ of the 1936 Olympics in Berlin and the 1938 FIFA
World Cup in Italy.
Mega-sporting events may also be used as a favourable
excuse to legitimise additional public spending that would not
otherwise pass through the political process. No example is
more pertinent than the looming London Olympic Games in
2012. Former London Mayor Ken Livingstone has even
brazenly admitted that he ‘didn’t bid for the Olympics because
[he] wanted three weeks of sport’, he bid ‘for the Olympics
because it’s the only way to get the billions of pounds out of
the Government to develop the East End – to clean the soil,
put in the infrastructure and build the housing’, succeeding in
his plot to ‘ensnare the Government to put money into an area
it has neglected for 30 years’.6 Given today’s economic climate,
the cost of hosting the Olympic Games in London has come
under increasing pressure from all spheres. It was only
recently that news broke that the government chose to ignore
a 250-page strategy document, signed off in December 2002
by Tony Blair as Prime Minister, that found little support for
the contention that the Games would produce significant
economic returns.7 Couple this with the embarrassing
admission by Olympics Minister Tessa Jowell who stated that
‘had we known what we know now’, London would not have
bid for the Olympics at all.8 The cost of hosting the 2012
extravaganza has increased from the anticipated £2.4 billion to
£9.35 billion.9 Indeed the government has already confirmed
they will be dipping into its contingency fund to make up for
the lack of private money for the building of the Olympic
village.10 Many are thus wondering at what cost the
government are pursuing what has been labelled by Stefan
Szymanski, Professor at the Cass Business School, as ‘a party,
which would be hugely expensive’.11


It is clear that mega-sporting events are extremely liable to
less-than-accurate sporting impact studies. These analyses may
overstate benefits, understate costs and misuse multipliers.
Opportunity cost remains a vital problem, but this has not
stopped events such as the Olympics becoming a new panacea
for economic and urban development. And while certain
benefits can be had from hosting sporting events, they are
accompanied by large caveats. New infrastructure must be
integrated into the host city’s economy and it must have a
clear legacy value. It is interesting to note that growth in the
number of impact studies conducted has coincided with a
surge in competition in the bidding process for these events.
The studies provide the backdrop for overzealous campaigning
and the acquisition of public money. Indeed, it is probably no
coincidence that most of the studies have been completed after
the 1984 Los Angeles Olympics, which was the first to make a
substantial profit.
1. Eric Barget considers a minimum threshold of a total audience of at least
one billion viewers, and/or 30 countries broadcasting the event.
2. Roads increased by 15% from those in 1986, new sewerage systems by
17% and new beaches and green areas by 78% (Brunet i Cid, 2002).
3. See
4. See
5. See
6. See
Mayor-tricked-Govt.-into-2012-Olympics-bid.html (25 April 2008).
7. See
(2 December 2008).
8. See
knew-about-recession.html (12 November 2008).
9. Ibid.
10. See
update_2012_budget_already_usi.html (22 January 2009).
11. See
Baade, R. (2006) ‘The Economic Impact of Mega-sporting Events’, in
W. Andreff and S. Szymanski (eds.) Handbook on the Economics of
Sport, Cheltenham: Edward Elgar.
Baade, R. and V. Matheson (2002) ‘Bidding for the Olympics: Fool’s
Gold?’ in C. P. Barros, M. Ibrahimo and S. Szymanski (eds.)
Transatlantic Sport: The Comparative Economics of North American
and European Sports, Cheltenham: Edward Elgar.
Baade, R. and V. Matheson (2004) ‘Mega-sporting Events in
Developing Nations: Playing the Way to Prosperity?’, South African
Journal of Economics, 72, 5, 1084–1095.
Barget, E. and J. J. Gouguet (2007) ‘The Total Economic Value of
Sporting Events Theory and Practice’, Journal of Sports Economics,
8, 2, 165–182.
Brunet i Cid, F. (2002) The Economic Impact of the Barcelona Olympic
Games, 1996–2004, Barcelona: Centre d’Estudis Olimpics UAB.
Coates, D. and B. Humphreys (1999) ‘The Growth Effects of Sport
Franchises, Stadia and Arenas’, Journal of Policy Analysis and
Management, 18, 4, 601–624.
Crompton, J. L. (1995) ‘Economic Impact Analysis of Sports Facilities
and Events: Eleven Sources of Misapplication’, Journal of Sport
Management, 9, 1, 14–35.
French, S. and M. Disher (1997) ‘Atlanta and the Olympics:
A One-year Retrospective’, Journal of the American Planning
Association, 63, 3, 379–392.
Greene, S. J. (2003) ‘Staged Cities: Mega-events, Slum Clearance, and
Global Capital’, Yale Human Rights and Development Law Journal,
6, January, 161–187.
Humphreys, B. and S. Prokopowicz (2007) ‘Assessing the Impact of
Sports Mega-events in Transition Economies: EURO 2012 in
Poland and Ukraine’, International Journal of Sport Management
and Marketing, 2, 5/6, 496–509.
Matheson, V. (2006a) ‘Economic Impact Analysis’, in W. Andreff and
S. Szymanski (eds.) Handbook on the Economics of Sport,
Cheltenham: Edward Elgar.
Matheson, V. (2006b) ‘Mega Events: The Effect of the World’s Biggest
Sporting Events on Local, Regional and National Economies’,
Working Paper No. 0622, International Association of Sports
Economists. Available at
Noll, R. and A. Zimbalist (1997) ‘The Economic Impact of Sports
Teams and Facilities’, in R. Noll and A. Zimbalist (eds.) Sports, Jobs
& Taxes: The Economic Impact of Sports Teams and Stadiums,
Washington, DC: Brookings Institution Press.
Owen, J. (2005) ‘Estimating the Cost and Benefit of Hosting Olympic
Games: What can Beijing Expect from its 2008 Games?’, The
Industrial Geographer, 3, 1, 1–18.
Pillay, U. and O. Bass (2008) ‘Mega-events as a Response to Poverty
Reduction: The 2010 FIFA World Cup and its Urban Development
Implications’, Urban Forum, 19, 3, 329–346.
Preuss, H. (2006) ‘The Olympics’, in W. Andreff and S. Szymanski
(eds.) Handbook on the Economics of Sport, Cheltenham: Edward
Siegfried, J. and A. Zimbalist (2002) ‘A Note on the Local Economic
Impact of Sports Expenditures’, Journal of Sports Economics, 3, 4,
Teigland, J. (1999) ‘Mega-events and Impacts of Tourism: The
Predictions and Realities of the Lillehammer Olympics’, Impact
Assessment and Project Appraisal, 17, 4, 305–317.
Whitson, D. and J. Horne (2006) ‘Underestimated Costs and
Overestimated Benefits? Comparing the Outcomes of Sports
Mega-Events in Canada and Japan’, in J. Horne and W.
Manzenreiter (eds.) Sports Mega-Events: Social Scientific Analyses of
a Global Phenomenon, Oxford: Blackwell, 73–89.
Jonathan Barclay is a former student at St Paul’s School, Barnes,
London, UK, and will take up a place to study at Yale University, USA,
in September 2009 (