Public Sector Support for Special Events

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/46526389

Public Sector Support for Special Events Article in Eastern Economic Journal · January 2009 DOI: 10.1057/eej.2009.31 · Source: RePEc

CITATIONS

READS

16

59

2 authors: Larry Dwyer

Peter Forsyth

UNSW Australia

Monash University (Australia)

223 PUBLICATIONS 4,021 CITATIONS

133 PUBLICATIONS 2,566 CITATIONS

SEE PROFILE

SEE PROFILE

All content following this page was uploaded by Peter Forsyth on 05 January 2017. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.

Eastern Economic Journal, 2009, 35, (481–499) r 2009 EEA 0094-5056/09 www.palgrave-journals.com/eej/

Public Sector Support for Special Events Larry Dwyera and Peter Forsythb a University of New South Wales, Sydney, NSW 2052, Australia. E-mail: [email protected] b Department of Economics, Monash University, Clayton, Victoria, 3800, Australia. E-mail: [email protected]

The standard approach taken to assessing the economic impacts of special events, using input-output (I-O) modeling, has come under increasing attack in the recent research literature. Critics have argued that both I-O modeling should be replaced by computable general equilibrium modeling and that cost benefit analysis should be used to account for the wider social and environmental impacts of events. This article explores the critics concerns, prior to highlighting the preferred economic impact assessment tool. The role of CBA is also discussed, highlighting the type of data that is needed for informed analysis. Researchers tend to treat economic impact analysis and CBA as distinct techniques of assessment, with the potential to provide conflicting recommendations. The article outlines a method for partially integrating the two techniques. To set a context for the discussion, reference is made to the findings from research recently undertaken to estimate both the economic impacts and the net benefits of the Formula One Grand Prix held in Melbourne, Australia. Eastern Economic Journal (2009) 35, 481–499. doi:10.1057/eej.2009.31 Keywords: special events; economic impact assessment; cost benefit analysis JEL: D5; D61; L83

INTRODUCTION There are sometimes good economic and non-economic reasons why a government may provide support for a special event. Special events increase the opportunities for new expenditure within a host region by attracting visitors to the region. They have the capacity to stimulate business activity, creating income and jobs in the short term, and generate increased visitation and related investment in the longer term. Sponsorship by governments of special events, even when they are run at a financial loss, is often justified by the claim that the events produce economic benefits for the region, and country, in which they are hosted. It is recognized that there may be other perceived benefits from events, such as enhancing the image of a city or region, facilitating business networking, and civic pride. Events can also result in associated social and cultural benefits to a destination, providing forums for continuing education and training, facilitating technology transfer, etc. On the other hand, events are recognized to generate adverse environmental impacts such as various forms of pollution and adverse social impacts such as disruption to local business and community backlash [Dwyer et al. 2000]. These aspects are very difficult to test or evaluate. Granted this, however, much of the public justification of events funding seems to center on their expected positive economic impacts, with the social and environmental effects treated separately.

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

482

There are two standard approaches to assessing events, the first being economic impact analysis, which estimates the impact of the event on variables such as Gross State Product (GSP) and employment. The other, cost benefit analysis (CBA), provides estimates of the wider effects of the event, both positive and negative, and attempts to put dollar values on these in order to estimate the overall result. The usual approach to event evaluation has been for researchers and consultants to estimate the economic impacts of an event, and then, alongside these, consider some of the possible wider effects of events that are not captured in the economic modeling. Effects of the latter type can be estimated using a formal CBA. This has resulted in a less than satisfactory approach to event evaluation, as the economic impact analysis and the CBA can give conflicting results. The economic impacts of special events are usually specified in terms of the multiplier effects on output and employment that are generated by expenditure on the event and by visitors. The multipliers are typically based on I-O modeling. In recent years, a number of articles have been published that have been critical of various aspects of this approach. Thus, critics such as Porter [1999]; Matheson [2002]; Matheson and Baade [2003]; Blake [2005] and Crompton [2006] argue that the economic impacts of events are often exaggerated. These critics have highlighted some inappropriate practices in event assessment that inflate the estimated economic impacts. These practices include exaggerating visitor numbers and expenditure, failure to deduct residents’ expenditure prior to modeling, abuse of multipliers, and inclusion of time switchers and casuals, as well as the tendency to ignore the various costs associated with special events (e.g. opportunity costs, costs borne by the local community, and displacement costs). These criticisms indicate that considerable caution must be employed before the results of any economic impact assessment of an event may be accepted. The issues that these critics raise certainly need to be better understood by the research community, but importantly they need to be better understood by industry stakeholders, as it is their hired consultants who often flaunt ‘‘best practice’’ in pursuit of their objective of providing the client with large numbers and optimistic economic impact assessments of events. In addition to the above types of ‘‘in-house’’ criticisms of event assessment, a number of articles and reports have appeared in recent years that have argued that the entire approach to event assessment needs a re-examination. One element of this approach emphasizes that the economic assessment models used for estimating the economic impacts of major events should reflect contemporary developments in economic analysis, particularly regarding the use of computable general equilibrium (CGE) modeling. More specifically, this has involved an attack on the uncritical use of I-O modeling, hitherto the standard technique of economic impact assessment of major events [Dwyer et al. 2005]. A second element in the new approach argues that event assessment that focuses only on economic impacts is too narrow in scope to provide sufficient information to policy makers and government funding agencies, and that, where practical, a cost benefit approach should be employed to embrace the emerging importance of social and environmental impacts in addition to economic impacts [Jago and Dwyer 2006]. CBA is the ‘‘ideal’’ approach to event assessment, as it is the expected net benefits from any use of government funds that should guide resource allocation. Unfortunately, CBA has enormous data requirements that make it rather impractical to apply for all events. Because of this, a review of the worldwide literature on event assessment reveals that, despite acknowledgement of the wider Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

483

social and environmental effects typically associated with events, only a handful of studies have employed a detailed CBA as an evaluation tool. This article has three main aims. The first is to highlight the limitations of the standard I-O technique in event economic impact assessment. As I-O analysis has inherent biases that overstate the impacts on output and jobs, it fails to provide information on industries adversely affected by the increased tourism demand. As a result, there is likely to be a misallocation of events funding and excessive overall spending in promoting events. In particular, I-O models will be contrasted with CGE models that provide potentially much more accurate assessments of an event’s economic impacts. The second aim is to argue that a focus solely on the economic impacts of events will provide insufficient guidance to policy makers as to whether or not the event warrants support by way of public funding. In particular, it will be argued that it is the expected net benefits of an event that is important in allocating scarce resources to its support. Thus, a more detailed assessment of a major event requires CBA. CBA is the most comprehensive of the economic appraisal techniques, and comprises a systematic process for identifying and assessing all (both direct and indirect) costs and benefits of an event, including, in principle though not always in practice, social and environmental costs and benefits. The third aim of the article is to address an issue about which there is much confusion on the part of tourism researchers and industry stakeholders. This relates to how the outcomes of economic impact assessment and CBA can be reconciled. Researchers tend to treat economic impact analysis and CBA as distinct techniques of assessment, with the potential to provide conflicting recommendations. Accordingly, this article will explore the relationship between the preferred economic impact assessment tool of CGE modeling, and CBA, highlighting the type of data that is needed for informed policy making. To set a context for the discussion, reference will be made to the findings from research recently undertaken to estimate both the economic impacts and the net benefits of the Formula One Grand Prix held in Melbourne, the capital city of the state of Victoria, Australia.

ECONOMIC ASSESSMENT OF EVENTS The multipliers used to estimate the impacts on output, income, and employment associated with a special event are invariably based on Input-Output (I-O) models [Burns et al. 1986; Crompton et al. 2001; Ryan and Lockyer 2001; Crompton 1995, 2006]. In recent years, a growing number of researchers have argued that I-O modeling is, in general, inappropriate for economic impact assessment of special events. Specifically, the concerns with I-O modeling relate to its rigid assumptions, such as exclusion of factor constraints and price changes, and neglect of industry interactive effects, which ignore economic realities and which act to exaggerate the economic impacts of events [Blake 2005; Dwyer et al. 2005; Jago and Dwyer 2006; Madden 2006]. Specifically, some limitations of the I-O approach for the purposes of event economic impact assessment are as follows:
No resource constraints exist for the inputs and resources supplied to the event. These resources are effectively assumed not to come from other industries, and their use in association with an event does not result in reductions in output elsewhere. Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

484


Prices and costs remain fixed, as economic activity expands due to the event. This excludes changes in factor and product prices with any consequent interactive effects that could lead to contraction of output and employment in other industry sectors.
There are constant proportions between inputs and output, between labor and output, and between value added and output. That is, any change in output of an industry supplying products and services associated with the event will lead to proportional changes in the quantities of its intermediate and primary inputs. The assumptions of I-O modeling imply that the injection of ‘‘new money’’ into an event location will always have a positive impact on production, income, and employment. The rigidity of these assumptions, however, is such that little confidence can be placed in the accuracy of the projected results. Unless there is significant excess capacity in tourism-related industries, the primary effect of an injection of expenditure into an area is to alter the industrial structure of the economy, rather than to generate a large increase in aggregate economic activity. Its effect will thus show up as a change in the composition of the economy, rather than as a net addition to activity. Key mechanisms, which determine the size of the economic impacts resulting from increased tourism demand, include factor supply constraints, real exchange rate appreciation, and current government economic policy [Dwyer et al. 2003]. The upshot is that I-O modeling does not provide an accurate picture of the economic impacts of events, and is thus incapable of informing event funding agencies of the ‘‘bangs for bucks’’ to be expected from event funding. Complementing the criticisms of I-O by researchers, there is now, in Australia and the UK at least, a growing critical literature by government agencies interested in developing the most robust and rigorous measures of event impacts [ACT AuditorGeneral 2002; Victorian Auditor-General 2007]. Federal and State government event funding agencies are increasingly demanding that economic impact assessment reports become more rigorous and transparent in terms of the economic models used to estimate economic effects such as changes to the GSP and employment and the rationale for key assumptions that have a material effect on the level of economic impacts. Recognizing the limitations of I-O models for event impact estimation, a growing number of researchers are employing CGE models for this purpose and to meet the rigorous constraints on event assessment demanded by funding agencies. CGE modeling allows for assumptions that better reflect economic realities such as the role played by factor constraints and changing prices on industry balance [McDougall 1995; Dixon and Parmenter 1996]. CGE models have a high degree of empirical content in the form of detailed commodity flows, labor market data, and national accounts data. They include more general specifications of the behavior of consumers, producers, and investors than those allowed in I-O models, thus permitting specific models to be calibrated to actual conditions in a particular economy. CGE models can estimate the impacts of increased event-related demand under a range of alternative macroeconomic scenarios, allowing for detailed interindustry analysis, together with supply side constraints and an active price mechanism. Space limitations preclude a discussion of the structure of CGE models, which, in any case, are tailored to a particular economy at a particular point in time. A good discussion of the structure of a CGE model as used in event assessment can be found in Blake [2005]. Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

485

CGE models provide a highly useful and flexible framework to examine the impacts of injected visitor expenditure associated with a special event. Events that have been assessed by using CGE models have included the Australian Formula 1 Grand Prix [Industry Commission 1996], the Rugby World Cup 2003 [URS Finance Economics 2004], Sydney 2000 Olympics [Madden 2006], and the London 2012 Olympics [Blake 2005]. For purposes of the economic impact assessment of special events, CGE models have several advantages over I-O models. They recognize that:
resource constraints on land, labor, and capital are generally present in an economy and that can limit changes in economic activity due to an event-related increase in the final demand for goods and services. For example, the events sector will need to expand output to meet increased demand by employing additional labor, land, capital, plant, and equipment. The constraints are perhaps most evident, however, in the case of labor that has some skills component. Some of these may be in limited supply, for example, specific labor skills or workers for particular shifts or locations.
relative prices may change due to an event, causing businesses to change the composition of their inputs. When there are capacity constraints, the prices on inputs and wages will increase in the face of an increase in demand, and the net impact of output and jobs from the increase in demand is much less than the initial injection of spending. These price rises will limit the extent of economic expansion associated with the event, and may even lead to contractions in economic activity in some sectors. With labor shortages, expansion in final demand associated with an event will increase the demand for labor, raising wage rates, and reducing the demand for labor elsewhere. Further, increased land values due to tourism development will impact the costs of other industries.
the behavior of the government budget sector has relevance for the estimated economic impacts of an increase in tourism expenditure associated with an event. For example, if additional infrastructure spending by government is required to support a special event, such as expenditure on stadia, roads, and airport landing facilities, there will be a positive effect on spending, but it must be financed. This may moderate the growth in private consumption associated with the event leading to downward pressure on the output of consumption-oriented industries. CGE modeling allows for assumptions that better reflect economic realities such as the role played by factor constraints and changing prices on industry balance [McDougall 1995; Dixon and Parmenter 1996]. CGE models have a high degree of empirical content in the form of detailed commodity flows, labor market data, and national accounts data. They include more general specifications of the behavior of consumers, producers, and investors than those allowed in I-O models, thus permitting specific models to be calibrated to actual conditions in a particular economy. CGE models can estimate the impacts of increased event-related demand under a range of alternative macroeconomic scenarios, allowing for detailed interindustry analysis, together with supply side constraints and an active price mechanism. Space limitations preclude a discussion of the structure of CGE models, which, in any case, are tailored to a particular economy at a particular point in time. A good discussion of the structure of a CGE model as used in event assessment can be found in Blake [2005]. For purposes of estimating the economic impacts of an injection of ‘‘new money’’ into an economy, such as might be associated with an event, CGE models can make specific assumptions about the availability of factors of production, such as to what Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

486

extent their supply can be increased, and to what extent there is an excess supply of some factors (as with unemployment of labor). The behavior of agents in a CGE model is assumed to be sensitive to changes in relative prices, as well as quantity variables. Price rises will limit the extent of economic expansion associated with the event, and may even lead to contractions in economic activity in some sectors [Powell and Lawson 1989]. Thus, for example, changes in land values due to event development will impact the costs of other industries. Where there are labor shortages, expansion in final demand associated with an event will increase the demand for labor, raising wage rates, and reducing the demand for labor elsewhere. It is an empirical matter as to how well integrated the state or regional labor markets in a particular country are [Partridge and Rickman 1998]. A CGE model can incorporate different assumptions as to how the labor market works, and illustrate the sensitivity of results to the different assumptions [Dwyer and Forsyth 1998]. CGE models produce more conservative results than I-O modeling, as they take into account capacity constraints and the impact of price changes (such as wages, rates of return on investment, and the exchange rate). An event brings additional demand to the economy, and as this demand is met, additional output and jobs are created. However, in real-life economies, when more resources are required in one area of the economy, they are drawn away, at least in part, from productive activities elsewhere in the economy [Dwyer et al. 2004; 2005; 2006a,b]. The prices of inputs and wages get bid up, and other activity is discouraged. Many of the impacts that I-O analysis ignores will thus be operating in the opposite direction from that of the initial spending boost. Despite the problems with I-O modeling of an event’s economic impacts and the existence of a superior technique, it is unfortunate that uncritical use of I-O modeling is still prevalent in the tourism literature.

CBA OF EVENTS Estimates of the economic impacts of an event provide, in themselves, an imperfect basis for decisions about resource allocation. Economic impacts, such as the change in GSP resulting from an event, are not the same thing as the economic benefits that arise. The change in GSP is a gross exaggeration of how much better off the state and, more precisely, its residents are. The measured impacts on economic activity of most tourism shocks, such as increases in tourism expenditure, may normally be expected to be much greater than the net benefits that they generate for the community (or in other words, the measure of the extent to which they make the community better off). CBA estimates the sum of welfare effects of an event for a particular community. These welfare effects include benefits and costs experienced by (a) consumers and producers of the event(s), and (b) other members of the community who may be neither consumers nor producers of these events but who, as third party participants, nevertheless share in the costs and benefits. The aggregate result of a CBA indicates whether the estimated gains exceed the costs to the community as a whole. If the estimated net social benefit of an event is positive (the total benefit exceeds the total cost), then the event is said to be an efficient use of society’s economic resources. The results of economic impact studies of a special event can only estimate the effect on the economy, whereas the question of whether an event is worth funding Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

487

can only be determined by estimating the event’s net benefit. A CBA can estimate the sum of welfare effects of a special event for a particular community. A welfare effect is simply any cost or benefit experienced by a member of the relevant community, consisting of consumers, producers, and third parties who are not directly involved in the event. In contrast, an economic impact study does not include benefits to consumers nor third party effects that are not captured in the markets as modeled. The benefits of an event outweigh the costs if the consumer’s willingness to pay outweighs the value of the resources used in the process (where value is represented by the opportunity cost involved). If the benefits exceed costs, there is a positive net social benefit, and the project should proceed. Conversely, if costs exceed benefits, there is a negative net social benefit and the project should be modified or not supported. The event literature comprises very few CBA of events. Recently a CBA of the forthcoming Winter Olympic Games in Vancouver 2010 was undertaken [Shaffer et al. 2003], where the authors forecast that the event will generate net costs to the Province of British Colombia of $1.2 billion. This is despite an economic impact study commissioned by the government of British Colombia that concluded that the event would generate claims by some proponents that the Games will generate over $10 billion in provincial Gross Domestic Product (GDP) and more than 200,000 jobs [InterVistas 2002]. Two post hoc studies undertaken on motor race events in Australia also conclude that the costs exceeded the benefits [ACT Auditor General 2002; Victorian Auditor-General 2007]. Each study aimed to provide an independent assessment of the economic value derived from a special event and to assist agencies in the selection and application of methodologies for evaluating major events. Interestingly, the Victorian Auditor General commissioned one independent consultant to apply CGE modeling to estimate the economic effects of the Formula 1 Grand Prix held in Melbourne, Victoria, in 2005, and commissioned another independent consultant to perform a CBA of the same event.

Outcome of CGE modeling The CGE model used to assess the economic impacts of the Grand Prix was the Monash Multiregional Forecasting (MMRF) model that has had a long history of use by various state and Australian government agencies, and is updated on an ongoing basis [Adams et al. 2003]. The model is built from the ‘‘bottom up’’ and states are linked via inter-state trade, inter-state migration, and capital movements. The MMRF model is comprised of eight Australian regions (the six states and two territories) and 56 sub-state regions (Table 1).

Table 1 Economic impacts of Australian Formula 1 Grand Prix, 2005 Region

Victoria Rest of Australia Nation

GSP/GDP $m

Private investment $m

Private consumption $m

Public consumption $m

Tax revenue $m

Employment

62.4 60.5

12.7 15.3

16.1 22.0

3.6 5.3

3.5 0.8

400

1.9

2.6

5.9

1.7

2.7

—

Current account balance $m

4.5

Source: Victorian Auditor-General [2007]. Dollars are $Australian (approx 65 cents US in January 2009). Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

488

The CGE simulations indicate that an increase in expenditure of $58.4 million injected into the State of Victoria as a result of the Grand Prix generates positive macroeconomic consequences, with real GSP up by $62.4 million, and real consumption increasing by $16.1 million. Real public consumption expenditure is assumed to move with real private consumption expenditure in response to the activity associated with the Grand Prix. The special event is found to generate 400 new jobs in the State (full-time equivalents). The Grand Prix crowds out activity in some industries both within the state and elsewhere in Australia. The modeling indicates that the event is associated with a substantial shift of resources and economic activity from the rest of Australia to Victoria, reducing GSP in other states by $60.5 million in total. Australian expenditures on the Grand Prix (that is, those by Victorians and people from other states) are funded wholly or partly by reductions in expenditures on other goods and services. In the case of inter-state visitors, this means that their spending in Victoria leads to equal-sized reductions in spending in their home states. This, combined with the negative effects of real appreciation on the competitiveness of traded-goods industries throughout Australia, is the main reason underlying the crowding out of activity in the rest of Australia, rendering national real GDP as essentially unaffected. Overall, due to a large decrease in GSP in other states, the Grand Prix generates a mere $1.9 million increase in GDP in Australia as a whole. The CGE model assumes that the Grand Prix has no effect on national employment, with the national real wage rate adjusting to ensure that national employment does not change. This is a standard long-run modeling assumption, based on the idea that in the long run national employment is determined by demographic factors (birth rates, death rates, etc.) and macro-economic management that are unaffected by the Grand Prix. Government budget balances (Federal and State) are held fixed in the model via endogenous changes in lump-sum payments to households. It follows, therefore, that any increase in taxation receipts arising from the Grand Prix will be immediately passed on to households. The fall in consumption outside Victoria leads to a fall in real national domestic ‘‘absorption’’ (the sum of consumption, investment, and government expenditure) relative to real gross domestic demand. Outcome of CBA Following the CBA of special events, the main types of benefits and costs may be identified. Benefits The benefits comprise payments to event organizers, resident consumer benefits, surpluses to state businesses, and surpluses to labor. (1) Payments to event organizers. For most events, the largest item of benefits will be the payments to the event organizer (e.g. ticket revenue, sponsorship revenue). The ticket sales and sponsor revenue received by the Australian Grand Prix Committee (AGPC) (exclusive of goods and services tax payments) are a benefit to the resident taxpayer, in that they offset the costs incurred in staging the Grand Prix, and reduce the size of the Victorian Government subsidy to the Grand Prix. Payments to organizers located outside the state are treated as costs. Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

489

(2) Resident consumer benefits may be classified into three types: (a) Consumer benefits (consumer surpluses) derived by resident householders who attend the event. Local residents, as patrons of the event, are willing to pay more to attend the event than the ticket price. Consumer surplus is the difference between the amount residents would be willing to pay for a ticket and what they actually pay. Ticket prices may not reflect the maximum that many consumers may be willing to pay for a good or service. The practice of ticket scalping is an indication that people are often willing to pay more than the official ticket price, and that consumer surplus exists. (b) Consumer benefits from attending related off-site events (satellite events). Many events have such activities associated with them. These usually comprise a mix of ticketed and non-ticketed activities. During the 2005 Grand Prix, most related events took place in Melbourne city center, including the parade of F1 cars and the live site in Federation Square attended by several thousand people. (c) Destination household benefits from indirect enjoyment of the event (nonattendees). This refers to the pride and enjoyment the residents would derive simply from hosting the event. (3) Surpluses to State businesses (returns to capital above opportunity costs). Business surplus is the difference between the value of output and the cost of the factors of production (land labor and capital), where their cost reflects their value in alternative uses. The source of business surplus comes when a firm is able to expand output, with the difference between the cost of the expanded output and the sale value being the business surplus. From this figure, we deduct income payable abroad (proportion of the surplus that accrues to out-of-state businesses). (4) Benefits to resident labor (above opportunity costs). An event may result in the creation of additional employment in the state or sub-region. Additional employment can be an extra source of net benefit over and above the business surplus. Labor surpluses are assumed to be generated because some jobs go to unemployed or underemployed workers who can be hired on a casual basis to meet the extra demand. The surplus accruing to labor is the amount of the wage above what is necessary to induce a worker to take on an employment opportunity, which in turn is determined by the wage in alternative employment opportunities. Thus, labor surplus occurs in the situation in which labor is employed at a wage higher than what workers would be prepared to accept to enter into employment.

Costs There are three main sets of costs of a special event. (1) Within the CBA framework, funds flow into the event organizer as benefits, and out as the costs of construction and operation of the event. Major categories of organizer costs include event management and staging, recurrent engineering, marketing/promotion and catering, and administration. These financial flows associated with an event can be estimated in a fairly straightforward way, as they are based on balance sheet figures of government departments and agencies. (2) Event-related costs incurred by other State government agencies. Other eventrelated costs incurred by other state government agencies might include payments in respect of roads and traffic authority, police, ambulance, fire brigade, and state emergency services. (3) Social and environmental costs include noise, congestion, disruption to resident lifestyles, traffic diversion, impact on destination image, carbon Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

490

footprint, and loss of access to amenities incurred by the host-destination community. The Grand Prix in Melbourne has for some years been the focus of concern for community groups that oppose the closure of suburban streets and parkland that is claimed to disrupt their lives. These effects, economic as well as non-economic, are often ignored, unless a particular effort is made to include them in a comprehensive event assessment. They represent, nonetheless, very real effects of events on a destination, and need to be recognized in the overall assessment of the costs and benefits of special events to the host destination. Many such effects are not sufficiently defined or measurable to be included in a CBA. As indicated in Table 2, the benefits of the Australian Formula 1 Grand Prix in 2005 were $63.1 million in total. The major component of benefits was visitor and sponsor payments to the AGPC. Minor sources of benefits comprised consumer surplus ($3.4 million), benefits of participants at related off-site events ($1.9 million), surpluses accruing to businesses ($3.7 million), and surpluses accruing to labor ($1.7 million). The costs of producing the 2005 Grand Prix event were estimated to be $69.8 million, while the estimated benefits were $63.1 million. This yielded an estimated net benefit (cost) of $6.7 million. The estimated net cost of $6.7 million is sensitive to assumptions made about Victorian consumer surpluses and Victorian business and labor surpluses. The latter depend importantly on the extent to which labor required to hold the event is diverted from other uses or from unemployment. The results of these sensitivity tests demonstrate that the size of the best estimate of the overall net benefit to Victoria from the Grand Prix does not change significantly in the face of plausible variations in the estimates for consumer surplus, business surplus, and labor surplus, and that none of the variations alone has the capability of turning the negative net benefit into a positive net benefit. Similarly, when the most optimistic and the most pessimistic of the sensitivity outcomes for consumer surplus, business surplus, and labor surplus are combined, the most optimistic outcome is $0.8 million and the most pessimistic outcome is $13.2 million [Victorian Auditor-General 2007, p. 122].

Table 2 Net benefit (cost) of Australian Formula 1 Grand Prix, 2005 Benefits K Visitor and sponsor payments to the AGPC $52.4 m K Victorian grand prix consumer surpluses $3.4 m K Other consumer benefits — e.g. benefits of participating in related off-site events $1.9 m K Surpluses gained by Victorian businesses $3.7 m K Surpluses gained by Victorian labor $1.7 m

$63.1 m

Costs K Construction and operating costs incurred by AGPC. $68.1 m K Grand prix-related costs incurred by other Victorian government agencies $0.5 m K Costs incurred by the Victorian community, including loss of Albert Park uses and amenity, traffic diversion and congestion, and noise $1.2 m

$69.8 m

Net Benefits

$6.7 m

Source: Victorian Auditor-General [2007]. Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

491

RECONCILING ECONOMIC IMPACT ANALYSIS AND CBA The results of the studies appear to point in different directions. The CBA study indicates that benefits fall short of costs by $6.7 million, while the CGE study indicates that the event lead to a $62.4 million increase in GSP, and generated 400 jobs. The CBA indicates that the Grand Prix is a poor investment, while the CGE study appears to suggest a highly positive outcome. Contrary to what might be implied in much of the economic impacts literature, estimates of the economic impacts of events provide, in themselves, an imperfect basis for decisions about resource allocation. The problem arises from a failure to distinguish clearly between the impacts and the (net) benefits of the event. The failure to make the distinction clear has, we believe, resulted in a situation in which tourism stakeholders generally regard ‘‘impacts’’ as synonymous with ‘‘benefits’’. ‘‘Impacts’’ on economic activity are measured by changes in GDP or similar measures. The change in GDP is an exaggeration of how much better off the country and, more precisely, its residents are when additional resources are used to enable this activity [Dwyer and Forsyth 1993]. ‘‘Net benefits’’ are a measure of the value of the gain in economic activity less the cost needed to enable this extra activity. It is interesting that the issue of reconciling economic impact analysis and CBA has received almost no attention in the research literature. Because both CBA and CGE are quite different in their purpose, method, and application, in the area of event assessment at least their results appear to be regarded as not directly comparable with one another. While a CBA addresses the extent of net social benefit to the host destination from the event itself, it cannot, however, measure the level of economic activity generated from the event or the wider flow-on effects. On the other hand, while economic impact analysis measures the level of economic activity associated with the event, it cannot address the issue of whether the event should receive government support. This is the view of the Victorian Auditor General [2007]. Consistent with this, the stance taken by those few researchers who have addressed the issue is that essentially they are complementary techniques, and one technique picks up aspects that are not picked by the other [Mules and Dwyer 2005]. In our view, treating the two techniques as distinct creates confusion among event managers and other stakeholders. In what follows, we explore an approach that attempts, as far as is possible, to integrate the two methods so as to better inform policy makers about the net benefits associated with a special event. Traditional CBA is primarily, though not necessarily, a partial equilibrium technique. It focuses on the direct impacts of a project, for example, the consumer surplus generated, the cash costs and revenues, the direct environmental externalities, and so forth. Shadow pricing normally seeks to correct for the direct distortions in input and output markets, for example, in handling taxes levied on fuel used by a project. However, the shadow-pricing problem is essentially a general equilibrium one — the shadow price of an input depends on how an extra unit of its use affects all markets in the economy, directly and indirectly [Boadway, 1975, and Dre`ze and Stern, 1987]. The general equilibrium nature of the cost benefit problem can be seen in the problem of shadow-pricing labor, often the most important input. If labor markets are perfect, and there are no taxes on labor, the market wage rate can be taken as the shadow price. However, if there is excess labor in the economy, or heavy taxation of labor, a shadow-pricing problem exists. An event will employ people, but it is not clear what net impact on employment and unemployment will result. To some Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

492

extent, the extra demand generated by the event will lead to wages being bid up, drawing workers away from other parts of the economy. Existing employees in affected sectors may work longer hours. If an event leads to additional employment and reduced unemployment in an economy, there will be a benefit, as the wages paid in event-related employment will exceed the shadow wage, or real cost of employment. Determining the impact on employment of an event and the cost of employing additional persons in the economy is essentially a general equilibrium problem. Partial equilibrium CBA also cannot handle situations in which the level of economic activity in a jurisdiction or national economy changes. One of the main benefits of events is that they stimulate economic activity in a region, state, or national economy. Determining the additional output, and use of inputs, is essentially a general equilibrium exercise, as the impact on activity, output, and the use of inputs depends on how markets across the economy work. In particular, they depend on how inputs markets work, such as for capital and labor. In a region or state, labor may flow in from other states, and if unemployment is present, the labor force can be augmented from unemployed persons. Even if unemployment is present or immigration is feasible, there may be limits on the availability of skilled employees, and increased demand may lead to higher wages rather than additional output. For the national economy, the potential labor force can be accepted as fixed for most countries, as immigration in response to labor market shortages is often not permitted. When there is additional expenditure in a national economy due to an event, the impact on economic activity will be moderated by exchange rate movements (induced by the rise in exports) and a reduction in output of other export- and import-competing industries. Typically, the positive impact of an event on economic activity will be larger in a state than in the national economy as a whole, as it will be possible to shift economic activity to the host state from other states — the supply of inputs to the national economy as a whole will be less flexible than to parts of the economy. As a result of stimulation of economic activity, welfare can increase. Additional economic activity leads to increased value added in the region (or GSP in a state or GDP in a nation). The extra output is valuable, but there are costs in producing it. Additional inputs, such as labor and capital, must be used to produce it, and these are costly. The net gain to the economy will be the value of the additional output less the cost of additional resources used. Even if new jobs are created, and unemployed persons gain employment, there is a cost, as they will have to forego leisure and will incur the disutility of work — typically, most unemployed persons would not be prepared to work for nothing. When employment in a state increases due to immigration from other states, the gain to the state will be small, as the net gain to the migrants is likely to be much smaller than the wage they receive once the costs of moving and the wages they could have obtained in the origin state are taken into account. Essentially, CBA is detailed but partial equilibrium, whereas CGE techniques are general equilibrium but less detailed. In principle, it would be possible to develop a CGE model that incorporates all the detail that one might include in a CBA, but in practice, this would be an impossibly demanding task. A CBA picks up a whole range of benefits and costs that would not be picked up in a CGE model. This includes non-priced effects that do not get included in the markets that are modeled — noise from an event, the consumer surplus of home patrons, loss of park amenity, and traffic congestion associated with the event. The CGE modeling in contract picks up the increased income to households resulting from the event. As neither Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

493

technique is completely comprehensive, both have a role in a comprehensive evaluation of a project or event. A similar conclusion was reached by Burgan and Mules [2001]. Their view can be illustrated by using a simple diagram. It is assumed that government funding to support the special event increases the demand for tourism services in Figure 1 from D1 to D2. The loss of consumer surplus is LRTJ, and the corresponding gain is KNRM. If both supply and demand are relatively elastic, the gain exceeds the loss. Producer surplus increases by JTML. Burgan and Mules note that, under CBA, the special event is worth undertaking if this gain exceeds the cost of running it. However, as the consumers of tourism services are mainly non-residents, they claim that consumer surplus is an inappropriate measure of economic or social gain from the point of view of the host economy. Considered in the context of Figure 1, the results of the economic impact analysis correspond with the increase in producer surplus JLMT, plus the increase in payments to factors of production TMXY. If the owners of the means of production are residents, this area represents extra income that the residents of the region receive as a result of hosting the event. Although it employs partial analysis only, the argument shows that the traditional economic impact measure has an appropriate theoretical correspondence with cost benefit analysis as a means of evaluating public expenditure on special events. On this basis, Burgan and Mules [2001] conclude that economic impact is to a degree consistent with the CBA perspective, in a situation entailing underused resources, suggesting that economic impact estimates of benefit and cost-benefit valuations can at broad levels be considered additive. This is an important conclusion, as it allows us to view consistently the range of possible benefits and costs involved. The use of a CBA approach, augmented with additional information obtained by using complementary techniques to fill in gaps in the CBA study, is used in other contexts. CBA is used extensively in the evaluation of transport projects. A recent Price K S N

L

M

R

J

T

D2 D1 Quantity Y

X

Figure 1. Impact of increase in tourism demand associated with an event [from Burgan and Mules 2001, Figure 2]. Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

494

British study in relation to evaluating transport projects [Eddington, 2006] endorses the use of CBA, but recognizes that a normal CBA may not capture all the relevant effects. Hence, it adds in the benefits from stimulation of GDP, obtained from modeling studies, to form a comprehensive evaluation of projects. The additional welfare benefits are significant though not enormous — for an illustrative group of road projects they add 5 percent to 45 percent to welfare benefits [Eddington 2006, p. 195]. One problem is that of overlap. Some aspects would normally be captured in both forms of evaluation. One of these would be the profit from an event. If an event earns a profit, this will be counted as a benefit along with consumer surplus and other benefits in a CBA. In a CGE study, if an event earns a profit, this should lead to an increase in real GDP, other things equal. When local patrons shift their expenditure from other goods and services to the event, the costs of the inputs used to create the event will be less than the value of its output. These resources can be used elsewhere to create other output of value, which will be added to GDP. A similar situation occurs when the event caters to visitors from outside the jurisdiction. If the price paid for the event by visitors is greater than the cost of the resources used to enable the event, then resources are freed to create value elsewhere, and GDP rises. In integrating the results of CBA and CGE studies, the rule should be to exclude any overlaps, or effects that are counted in both studies. The cost of raising taxes to fund a subsidy is something that can also be handled in both the CBA and CGE analysis. In a CBA, if taxes are raised to subsidize the event, there will be a cost due to the deadweight loss created from raising taxes. Raising $1 of tax imposes more than an additional $1 of cost on the economy because the additional tax increases an already present distortion. Taxes cannot be raised costlessly. It is estimated that the cost of raising $1 of tax in Australia is $1.2–$1.5 (for discussion, see Freebairn [1995]; Campbell and Bond [1997]), and in a state of Australia more than this, because state governments only have access to relatively distorting taxes. If an event incurs a loss to be financed by government, there is an additional cost that needs to be taken into account. This cost is automatically taken account of in a CGE, but must be explicitly addressed in a CBA. In a CGE analysis, the additional taxes required to fund the subsidy would lead to a shift in the composition of output (even in a full employment situation in which there is no macro effect of the tax on economic activity). The output of taxed goods with a high price to cost ratio would be lowered to increase the output of a low price to cost ratio, and this will lead to a reduction in real GDP. In addition, a rise in taxes in a state may make the state less competitive, which would lead to a shift in economic activity to other states. This additional effect may not matter for the national economy as a whole, but it would raise the cost of funding subsidies for the event’s host state.

INTEGRATION IN PRACTICE In fact, the CBA study does include some benefits that come about through indirect effects throughout the economy, and these are measured by using CGE techniques. A CBA would detail benefits and costs, such as consumers’ surplus, which would not be picked up in a CGE study, while the CGE study would pick up general equilibrium effects that the partial CBA is not capable of detecting. Some CGE models are explicitly designed to measure changes in welfare [Dixon et al. 1982]. In his study of the impacts of the London Olympics 2012, Blake [2005] includes a measure of resident welfare. Consistent with economic theory, Blake’s Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

495

model measures a change in welfare by equivalent variation (EV), which indicates how much the change in welfare is worth to the economy at the pre-simulation set of prices. This measure takes the results from what may be quite complex effects of a simulation on a household, and produces a single value to describe how much better (or worse) off the economy is as a result of such effects. Blake’s estimates employ EV, along with production measures such as Gross Value Added (GVA) and GDP, to assess the welfare effects of simulations on the economy [Blake 2005]. The credibility of such estimates depends upon how robust they are to alternative specifications of the utility functions, production functions, ability of factors to move from jurisdiction to jurisdiction, and many other assumptions inherent in the CGE model. One step toward assessing the gain to the economy from the increase in economic activity is to measure the changes in surpluses (or differences between gross gains and the costs of obtaining them). As we have indicated, owners of capital/businesses, governments, and workers may gain from increased economic activity. Moreover, if tax receipts increase as a result of increased economic activity, there is a surplus gain that accrues to the government. Following the estimates of the economic impacts of an event on GSP, two measures of ‘‘net benefit’’ can be derived, namely, business surplus and labor surplus, both of which were defined above. Dwyer et al. [2006b] have shown how business surplus can be derived from the outputs of a CGE model. When an event is held, businesses such as hotels and restaurants and tour companies are likely to accrue business surpluses because of some likely spare capacity. Thus, for the Grand Prix, if local businesses were able to accommodate the extra demand created by inter-state and international visitors to the event without reducing their services to other customers, and without requiring additional capital and land, then those businesses would earn surpluses. In the approach of Dwyer et al. [2006b], a measure of ‘‘business surplus’’ is generated by subtracting the cost of additional labor used (wage by quantity), the cost of additional capital services supplied domestically, and cost of additional natural resources from the change in the value of the increased economic activity, as measured by the change in GSP. From this figure, income payable abroad (proportion of the surplus that accrues to out-of-state businesses) must be deducted. Used for event impact analysis, a public sector agency could employ a CGE model to estimate the impacts of an event on GSP, and then deduct the costs of labor from this figure to derive the value of business surplus. Adopting this approach for the Grand Prix, as indicated above, business surplus is estimated to be $3.7 million. The estimation of surpluses to labor is slightly more complex. The standard procedure in CBA is to assume full employment, so that all labor, capital, and land are valued at their opportunity cost. There are circumstances, however, in which an event may well result in the creation of additional employment in the state or subregion. Any additional employment can be an extra source of net benefit over and above the business surplus. Labor surpluses are generated when jobs go to unemployed or underemployed workers, who can be hired on a casual basis to meet the extra demand. This is often the situation for a special event. A labor surplus occurs in a situation in which the additional labor is employed at a wage higher than what workers would be prepared to accept to enter into employment. Thus, the surplus accruing to labor is the amount of the wage above what is necessary to induce a worker to take on an employment opportunity, which in turn is determined by the wage in alternative employment opportunities. Additional jobs due to the holding of events would bring a positive benefit to the economy, as persons who are Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

496

involuntarily unemployed would be willing to work for less than they will be paid if they obtained a job. The difference between the cost of labor evaluated at the market wage rate and evaluated at the shadow or reservation wage is an additional benefit from the increased economic activity associated with an event. The CGE model can allow for the effects of different costs of expanding the amount of labor employed. If additional event-related expenditure leads to less involuntary unemployment, additional labor should be costed at less than the market wage, and the measured benefit from the event should be larger. With information on the shadow wage, the labor surplus can be calculated and added to the measure of net benefits produced by the CGE model. In an earlier attempt to estimate labor surplus from simulations by using a CGE model, Dwyer et al. [2005] assumed that workers who gained jobs because of an increase in tourism demand were willing to work for two-thirds of the wage that they received. This implied a surplus for each worker equal to one-third of the wage actually received. The overall labor surplus was thus estimated to be one-third of the additional wages bill associated with the additional employment. As wage rates are known (from estimation of the business surplus), the labor effect can then be derived in a straightforward manner. The same assumptions underlying the study by Dwyer et al. [2005] were employed by the Victorian Auditor General [2007] to estimate the labor surplus figure of $1.7 million used in the CBA of the Grand Prix. If it is assumed that all of the output reflected in the value of GSP is produced by the application of unemployed or part-time labor to the spare capacity available in capital and land, all of the difference between labor cost and GSP is a surplus that accrues to existing capital and land. This is because no additional capital and land were required to meet the extra demand. The Victoria Auditor General adopted this approach to estimate business surplus as part of the CBA of the Grand Prix in Melbourne [Victorian Auditor-General 2007]. The measures of business surplus and labor surplus can be added together to comprise a measure of the ‘‘net economic benefit’’ of an event that a government can employ in its assessment of which degree of support, if any, is warranted for a special event. In summary: Impact of event on real GSP Fewer costs of the factors of production (land, labor, and capital) ¼ Business surplus þ Labor surplus ¼ Net Economic Benefit of Event It must be stressed that business surpluses and labor surpluses do not comprise the total net benefits — they form only part of a CBA. As discussed, costs to government need to be deducted, whereas benefits to consumers, and benefits to residents, including effects on third parties, need to be added to the estimates of business and labor surpluses to provide a more comprehensive estimate of the net (overall) net benefits of an event. It is also recognized that there are various other event-associated costs and benefits that are not included in any of the above measures. These items include various wider economic effects, such as future increased tourism numbers and expenditure, increased investment, enhanced trade and business development, as well as social and environmental effects that are difficult to measure but must, nevertheless, be acknowledged in any comprehensive accounting of the costs and benefits of an event. Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

497

CONCLUSIONS This article had three aims. The first was to highlight the limitations of the standard I-O technique in event economic impact assessment. We have argued that the standard approach, using multipliers from I-O models, is likely to result in a misallocation of events funding and excessive overall spending in developing events. CGE models are the preferred technique for assessing the impact of an event on economic activity, and its various dimensions such as GSP/GDP and employment. The CGE approach gets around the serious problems encountered by I-O techniques, so often used (or misused) in event evaluation, for assessing impacts. The second aim was to argue that a focus solely on the economic impacts of events provides insufficient guidance to policy makers as to whether or not the event warrants support by way of public funding. To do this, we contrasted the results of an economic impact assessment and a CBA of the same event. To illustrate some results of CGE and CBA analysis of the same event, we referred to two studies recently commissioned by the Victoria Auditor General as part of a comprehensive study of best practice event assessment [Victorian Auditor-General 2007]. The CBA study indicates that benefits fall short of costs by $6.7 million, whereas the CGE study indicates that the event lead to a $62.4 million increase in GSP, and generated 400 jobs. The CBA indicates that the Grand Prix is a poor investment, whereas the CGE study appears to suggest a highly positive outcome. The third aim of the article was to explore how the outcomes of economic impact assessment and CBA can be reconciled. We have argued that there is indeed common ground between CBA and economic impact analysis of special events. In particular, it was argued that the two techniques focus on different aspects of the evaluation problem. CBA is the established technique for assessing the wider benefits and costs of a project, and as such, it is an appropriate framework for classification and measurement of the projected outcomes of an event. While CBA would emphasize consumer surplus as a primary source of gains from an event, in the case of larger special events, their outcomes are not aimed specifically at providing benefits to local consumers, but rather at attracting tourists and their expenditure from outside the region. It would seem then that economic impact analysis can provide important information on what is essentially a major source of benefit of a special event. A complete integration of the results of the two studies outlined above was not feasible given the information provided. However, we identified how the different elements can be integrated, giving particular attention to the problem of overlap. Recognizing that changes in GSP/GDP are gross changes, and not measures of the net welfare gain, we noted some of the benefits and costs arising from changes in economic activity as estimated in the CGE simulation. When the subsidy costs of achieving these changes are taken account of, the Grand Prix looks more marginal, and the assessment is consistent with that of the CBA study. The use of CGE modeling to derive business and labor surpluses that comprise important components of a CBA represents an important step in clarifying and reconciling the differences that often exist between impact analysis and CBA of a special event. The approach that we recommend in this article bridges the gap between economic impact analysis and CBA, in a way that has policy relevance for a public sector agency in order to inform its decisions in respect of which events should receive support and which events should not be supported by government funds. The effectiveness of this approach to evaluating the impacts of special events Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

498

will, of course, always be constrained by the availability of data. As indicated, the concept of ‘‘net economic benefit’’ excludes many effects of events that need to be taken into account in any comprehensive event assessment. Important components in estimating costs and benefits, such as those relating to environmental and social impacts, can be inherently difficult to value. This will also apply to any ‘‘feel good’’ component that the hosting of an event may provide for a community. The perceived impact on future tourism or investment flows as a result of wide media coverage, or from the business networking opportunities provided, have also been used to justify the hosting of a major event or even for the cumulative effect on a city or region’s wider profile from hosting a regular ongoing series of events. Evaluations of these effects remain fertile ground for further research.

Acknowledgements The research for this article has been supported by the Sustainable Tourism Cooperative Research Centre, Gold Coast, Australia. The authors acknowledge the critically constructive comments of the journal referees, which have led to improvements in the article.

References ACT Auditor General. 2002. ACT Auditor General’s office Performance Audit Report V8 Car Races in Canberra — Costs and Benefits. Canberra, ACT: Auditor General Department. Adams, P., M. Horridge, and G. Wittwer. 2003. MMRF-GREEN: A Dynamic Multi-Regional Applied General Equilibrium Model of the Australian Economy Based on the MMR and MONASH Models, CoPS Working Paper G-140, Melbourne: Centre of Policy Studies and Impact Project, Monash University, 70. Blake, A. 2005. The Economic Impact of the London 2012 Olympics, TTRI Discussion Paper 5/2005, Christel DeHaan Tourism and Travel Research Institute Nottingham University Business School. Boadway, R. 1975. Cost Benefit Rules in General Equilibrium. Review of Economic Studies, 42(July): 361–373. Burgan, B., and T. Mules. 2001. Reconciling Cost — Benefit and Economic Impact Assessment for Event Tourism. Tourism Economics, 7(4): 321–330. Burns, J.P.A., J.H. Hatch, and T.J. Mules. 1986. The Adelaide Grand Prix: The Impact of a Special Event. Adelaide: Centre for South Australian Economic Studies. Campbell, H., and K. Bond. 1997. The Cost of Public Funds in Australia. Economic Record, 73(March): 22–34. Crompton, J.L. 1995. Economic Impact of Sports Facilities and Events: Eleven Sources of Misapplication. Journal of Sport Management, 9: 14–35. _______ . 2006. Economic Impact Studies: Instruments for Political Shenanigans? Journal of Travel Research, 45: 67–82. Crompton, J., S. Lee, and T. Shuster. 2001. A Guide for Undertaking Economic Impact Studies: The Springfest Festival. Journal of Travel Research (1): 79–87. Dixon, P., and B. Parmenter. 1996. Computable General Equilibrium Modelling for Policy Analysis and Forecasting, in Handbook of Computational Economics, 1, edited by H. Aman, D. Kendrick and J. Rust Amsterdam: Elsevier Science B.V., 4–85. Dixon, P., B.R. Parmenter, J. Sutton, and D. Vincent. 1982. ORANI-F: A Multisectoral Model of the Australian Economy. Amsterdam: North Holland Publishing Co. Dre`ze, J., and N. Stern. 1987. Theory of Cost Benefit Analysis, in Handbook of Public Economics II, edited by A Auerbach and M. Feldstein Amsterdam: North Holland. Dwyer, L., and P. Forsyth. 1993. Assessing the Benefits and Costs of Inbound Tourism. Annals of Tourism Research, 20(4): 751–768. _______ . 1998. Estimating the Employment Impacts of Tourism to a Nation. Tourism Recreation Research, 23(2): 3–12. Eastern Economic Journal 2009 35

Larry Dwyer and Peter Forsyth Public Sector Support for Special Events

499 Dwyer, L., P. Forsyth, and R. Spurr. 2004. Evaluating Tourism’s Economic Effects: New and Old Approaches. Tourism Management, 25: 307–317. _______ . 2005. Estimating the Impacts of Special Events on the Economy. Journal of Travel Research, 43: 351–359. _______ . 2006a. Assessing the Economic Impacts of Events: A Computable General Equilibrium Approach. Journal of Travel Research, 45: 59–66. _______ . 2006b. Assessing the Economic Impacts of Special Events, in International Handbook of Tourism Economics, edited by L. Dwyer and P. Forsyth England: Edward Elgar. Dwyer, L., P. Forsyth, R. Spurr, and T. Ho. 2003. Estimating the Regional and National Economic Impacts of Tourism Growth and Special Events. ASEAN Journal of Hospitality and Tourism Management, 2(2): 3–19. Dwyer, L., R. Mellor, N. Mistilis, and T. Mules. 2000. A Framework for Assessing ‘‘Tangible’’ and ‘‘Intangible’’ Impacts of Events and Conventions. Event Management, 6(3): 175–191. Eddington, R. 2006. The Eddington Transport Study: Transport’s Role in Sustaining the UK’s Productivity and Competitiveness. London: HMSO. Freebairn, J. 1995. Reconsidering the Marginal Welfare Cost of Taxation. Economic Record, 71: 121–131. Industry Commission. 1996. State and Local Government Assistance to Industry; Report 55. Canberra: Australian Government Publishing Service. Jago, L., and L. Dwyer. 2006. Economic Evaluation of Special Events: A Practitioners Guide. Altona, Australia: Common Ground Publishing Pty. Ltd. InterVistas. 2002. The Economic Impact of the 2010 Winter Olympic and Paralympic Games: An Update. Vancouver, British Columbia: Minister of State for Community Charter and 2010 Olympic Bid, http://www.mcaws.gov.bc.ca/2010/Econ_Impact_2010_Games_Update.pdf (accessed October 11, 2005). Madden, J. 2006. Economic and Fiscal Impacts of Mega Sporting Events: A General Equilibrium Assessment. Public Finance and Management, 6(3): 346–394. Matheson, V. 2002. Upon Further Review: An Examination of Sporting Event Economic Impact Studies. The Sport Journal, 5(1): 1–3. Matheson, V., and R. Baade. 2003. ‘‘Bidding for the Olympics: Fools Gold?’’ in Transatlantic Sport, edited by C. Barros, M. Ibrahim and S. Szymanski London: Edward Elgar Publishing. McDougall, R. 1995. Computable General Equilibrium Modelling: Introduction and Overview. AsiaPacific Economic Review, 1: 88–91. Mules, T., and L. Dwyer. 2005. Public Sector Support for Sport Tourism Events: The Role of CostBenefit Analysis. Sport in Society, 8(2): 338–355. National Institute of Economic and Industry Research. 2005. Economic Impact Evaluation of the 2005 Foster’s Australian Grand Prix, A Report for the Australian Grand Prix Corporation, Melbourne. Partridge, M.D., and D.S. Rickman. 1998. Regional Computable General Equilibrium Modeling: A Survey and Critical Appraisal. International Regional Science Review, 21: 205–248. Porter, P. 1999. Mega-Sports Events as Municipal Investments: A Critique of Impact Analysis, in Sports Economics: Current Research, edited by J.L. Fizel E. Gustafson and L. Hadley. New York: Praeger Press. Powell, A., and A. Lawson. 1989. A Decade of Applied General Equilibrium Modelling for Policy Work, in General Equilibrium Modelling and Economic Policy Analysis, edited by Bergman L, Jorgenson D. and Zalai E. New York: Blackwell, 241–290. Ryan, C., and T. Lockyer. 2001. An Economic Impact Case Study: The South Pacific Masters Games. Tourism Economics, 7(3): 267–275. Shaffer, M., A. Greer, and C. Mauboules. 2003. Olympic Costs and Benefits. Vancouver: Canadian Centre for Policy Alternatives Publication. URS Finance Economics. 2004. Economic Impact of the Rugby World Cup 2003 on the Australian Economy-Post Analysis, A Report for the Commonwealth Department of Industry Tourism and Resources, Canberra, June. Victorian Auditor General. 2007. State Investment in Major Events. Victoria: Victorian Government Printer, May.

Eastern Economic Journal 2009 35

View publication stats