Skip to content

How to evaluate sport and culture: 1996 Summer Olympics, Atlanta GA (statistical approach)

arrow down
kiarash-mansouri-E3cvT5MzGZ0-unsplash

What was the programme and what did it aim to do?

This study evaluates the employment impacts of the 1996 Summer Olympics in Atlanta, Georgia. As with most sports and culture projects, the objectives were not primarily about delivering local economic benefits. But this event was expected to generate significant ‘legacy’ impacts, particularly employment gains. Such a jobs boost might come directly (via spending on the events and Games infrastructure, or through visitor spending), or indirectly (through job training provided to local workers and unemployed people). Conversely, the legacy might be negative if Olympic-related expenditure diverted spending away from other job-creating activities; or if some potential visitors stayed away because of the games (or locals left town).

What’s the evaluation challenge?

Identifying the employment effects of a mega-event like this is not easy because only specific types of cities tend to host these events. Winning cities are selected by a committee, who may favour cities with certain characteristics. For example, it may be that only cities with a good transport infrastructure are chosen to host the event. As a result of this selection, if we compare differences in outcomes for host cities to those in other cities, these differences may not reflect the impact of the event. Instead, they may simply reflect differences in the types of city who hosts these events.

What did the evaluation do?

To deal with these issues, the authors compared changes in employment outcomes in treatment and control areas using a difference in difference approach. Treatment areas are counties containing Olympic venues, or those immediately around those counties. Control areas are counties elsewhere in Georgia that were similar to treated counties based on their observable characteristics. The similar control counties were selected using a method known as matching.

How good was the evaluation?

According to our scoring guide, matching combined with difference-in-differences receives a maximum of 3 (out of 5) on the Maryland Scientific Methods Scale (Maryland SMS). This is because it does well to control for observable differences (e.g. productivity) between event hosting and non-event host areas, but is unable to control for unobservable differences (e.g. the quality of local infrastructure). Since this study uses a wide range of variables in its matching and since the difference-in-difference is based on a clear treatment date we score this study 3 on the Scientific Maryland Scale.

What did the evaluation find?

This study finds no significant effects of the Games on county employment growth. The ‘Olympic counties’ were already experiencing strong employment growth, to which the Games added very little. Allowing the Olympic effect to start from 1994, the study finds a positive (weakly significant) effect, adding up to a 1% boost to jobs growth in Olympic counties. They also find a weak ‘rebound’ effect in Olympic counties between 1995 and 1996 (although technically these results are insignificant meaning the effects are difficult to detect and may be zero). Overall, there is very little evidence for an Olympic employment legacy.

What can we learn from this?

What can we learn from these results? Modern bids for Olympic Games almost always highlight a ‘legacy’, which typically includes positive employment effects. However this study suggests that such effects may have been zero for the Atlanta games. If this result generalises (as suggested by our evidence review) then mega-events like London 2012 and Glasgow 2014 may find similarly small employment impacts. It also suggests that smaller sporting events and festivals – which typically don’t involve new infrastructure – are unlikely to generate long term net job gains. That does not mean we shouldn’t run or fund such events, of course – simply that the main benefits are unlikely to be economic.

Reference

Feddersen, A., and Maennig, W. (2013a). Employment Effects of the Olympic Games in Atlanta 1996 Reconsidered. Int. J. Sport Financ. 8, 95–111. [Study 321 from our Sports and Culture review]

How to evaluate sport and culture: 1996 Summer Olympics, Atlanta GA (statistical approach)