« Illegal immigration and crime | Main | Are Good Times, Bad Times? »
Trade and the signaling benefits of hosting the Olympics
Andrew Rose and Mark Spiegel report:
Economists are skeptical about the economic benefits of hosting “mega-events” such as the Olympic Games or the World Cup, since such activities have considerable cost and seem to yield few tangible benefits. These doubts are rarely shared by policy-makers and the population, who are typically quite enthusiastic about such spectacles. In this paper, we reconcile these positions by examining the economic impact of hosting mega-events like the Olympics; we focus on trade. Using a variety of trade models, we show that hosting a mega-event like the Olympics has a positive impact on national exports. This effect is statistically robust, permanent, and large; trade is around 30% higher for countries that have hosted the Olympics. Interestingly however, we also find that unsuccessful bids to host the Olympics have a similar positive impact on exports. We conclude that the Olympic effect on trade is attributable to the signal a country sends when bidding to host the games, rather than the act of actually holding a mega-event. We develop a political economy model that formalizes this idea, and derives the conditions under which a signal like this is used by countries wishing to liberalize.
An ungated copy of the paper is to be found here (which, I should note, I have not read).
Could it be unobserved heterogeneity, namely that up-and-coming nations (and there is no perfect way to control for that) are the ones who apply to host the Olympics? It costs a lot of money to put together a bid and Albania is simply not going to try. It is hard for me to believe that unsuccessfully submitting a bid to host the Olympics should boost exports by 30%. The simpler model is that winner's curse holds, overbidding for the games results, bidding is also driven by rent-seeking and special interests, but since the bidders are up-and-coming cities in the final analysis the complaints don't sound so convincing.
Posted by Tyler Cowen on April 14, 2009 at 05:54 AM in Sports | Permalink
Comments
Countries which bid for the Olympics are probably similar in that they are all dynamic, out-going countries. However, the winning bid receives far more attention, and should increase exports much more. That it does not suggests that the heterogeneity point is dominant.
Posted by: Thorfinn at Apr 14, 2009 4:45:54 AM
Research shows that eating in Michelin-starred restaurants makes you rich. Even trying, but failing, to reserve a seat makes you rich.
Posted by: Zamfir at Apr 14, 2009 4:57:58 AM
It's all very well to analyse winning the bid to host the Olympics, but what about winning actual Olympic medals?
This is going back a while now, but I wrote a post during last summer's Olympics trying to work out the economic effect of Britain's surprisingly large haul of gold medals.
http://www.knowingandmaking.com/2008/08/economic-efficiency-of-sport.html
Posted by: Leigh Caldwell at Apr 14, 2009 7:00:43 AM
This really sounds a lot like your observation on Nobel laureates: unobservable heterogeneity if you like; I'd call it old-fashioned selection bias. Problem is, there probably isn't a sufficient supply of control cases (up-and-comers who are active in the international relations/trade arenas, who are also going to have booming trade anyway simply because they're about to reform from or in the process of reforming Maoist/Leninist systems, etc) to compare. And I'd like to see the results controlled for the economic environment, too -- an event that happens every four years is bound to hit one side or another of the global cycle more frequently than the others.
WHat I'd also like to see is what the result would be of a non-bid that involved extensive development spending on infrastructure and tourism revenue-generating facilities from an equivalent country.
Posted by: Tom at Apr 14, 2009 7:42:48 AM
Atlanta and Salt Lake City saw pretty good growth following the olympics. How large that would have been without them I have no idea. SLC area has had some pretty large foreclosure numbers.
Havanna will make a bid next go round.
Posted by: oops at Apr 14, 2009 7:59:48 AM
Atlanta got a surge of both population and home building/home prices. The Olympics was a key step in the transition to more infill development and downtown housing. Ironically many of the Olympic-funded projects were disasters (new dorms in the wrong location for Georgia State, new dorms at Georgia Tech were trashed by athletes and had to get renovated right away, open-air pool had to be enclosed and had lots of problems), but they did encourage private developers to build more in the area.
It is of course impossible to distinguish the Olympic effect from what would have happened later, but it did give the city a better image and it did affect the location of development within the city.
Posted by: DK at Apr 14, 2009 8:33:37 AM
nice one, zamfir.
Posted by: andrew at Apr 14, 2009 8:43:12 AM
I guess that the best strategy is to go through the process of bidding to host, but coming up short. You get the benefits without having to build all of that infrastructure that will have little post-Olympics use (velodromes anyone?). Of course, this requires that the unsuccessful bidder have a credible claim to be an Olympic host.
Posted by: Eric at Apr 14, 2009 9:01:35 AM
30% sounds unbelievable.
It may be right, of course, every now and then science does show up incredible results that are true. But a 30% increase makes me prickly.
And really how does intending to host Olypmic games signal an intention to liberalise? I can understand why a PhD signals an ability to work long hours and delay gratification, but the link between liberalising and hosting Olympic games doesn't seem nearly so obvious.
And why is it in policy-makers' interests in the central government and the big city to liberalise?
Posted by: Tracy W at Apr 14, 2009 10:07:40 AM
As always, correlation is not causation.
Posted by: anonymous at Apr 14, 2009 10:26:17 AM
Oh, man, I hope it's true at a local level too:
"Birmingham, Alabama, mayor wants Olympics in 2020"
http://www.al.com/news/birminghamnews/index.ssf?/base/news/1214036114220520.xml&coll=
Patrick
Birmingham, AL
Posted by: Patrick at Apr 14, 2009 11:03:14 AM
As always, they are not stupid and aware of correlation is not causation. They tried to adjust for endogeneity, and the results seem to be robust. Of course their paper is not necessarily the last word, and we cannot be absolutely sure that their adjustement for endogeneity was truly successful.
Posted by: Severin at Apr 14, 2009 11:14:22 AM
As if many people are aware of who bid and failed. There is no mechanism for this to happen.
Posted by: stanfo at Apr 14, 2009 12:34:40 PM
Surprised not to see mention of Montreal here.
The "jobs in Atlanta" argument reminds me of casino studies. IIRC, these show jobs are created only in the immediate vicinity of the casino (where they are easy to measure), but more are lost elsewhere (where they are diffused, and harder to measure).
Posted by: Zbicyclist at Apr 14, 2009 2:13:37 PM
Everyone seems to agree with Tyler's criticism, but I don't understand the difference between the paper's conclusion and Tyler's criticism.
Does it hinge on whether a hypothetical non-Olympic-bidding but still "liberalizing" country would not see the same economic gains as a bidding-and-liberalizing country? If so, what are some of those countries?
If there aren't any, the criticism and the paper seem like they are saying the same thing: Liberalization is good for the economy and liberalizing countries are more likely to bid on the Olympics.
Posted by: Curt Fischer at Apr 14, 2009 5:01:49 PM
Obama Lends Weight to World Cup Bid
http://www.nytimes.com/2009/04/15/sports/soccer/15vecsey.html?_r=1&ref=sports
Way to boost exports...
Posted by: Cocoa at Apr 14, 2009 8:01:58 PM
Well,
So the best policy any government (even the albanian one) should do is to bid to every major event in the world (once that it doesn't matter if you win or not).
I think that you cannot call this type of thing as signalling. Once all country types can bid for any event (there no cost at all in doing this. You can even say that this is not risky, once that the countries which would be hurted if hosting these kind of events will never be picked to host them).
Posted by: Claudio at Apr 15, 2009 4:01:32 PM
Severin,
Have you looked at their strategy for addressing the selection bias? They don't control for trade policy, which suggests at first glance that Tyler's skepticism is warranted.
http://www.tradediversion.net/archives/2009/04/i-doubt-the-olympic-effect.html
Posted by: Jonathan at Apr 16, 2009 9:34:15 AM
Curt - the paper introduces a further hypothesis - that cities bidding for the Olympics are signalling that they are committed to liberalising. Why bidding on the Olympics would signal a commitment to liberalising is not discussed in the paper and I can't figure out a plausible explanation.
Posted by: Tracy W at Apr 17, 2009 7:26:55 AM
