Johan Fourie's blog

I'd rather be a comma than a fullstop

World Cup winners

with 3 comments

(c) Johan Fourie

Sea of yellow: Watching Bafana play France was one of my most memorable sporting moments

Sunday’s Premier League final day could not have been scripted more dramatic. It was a good day to be an Arsenal supporter, but a once-in-a-lifetime experience to be a Man City fan. It was a match that awakened bipolar emotions: the joy of a first goal and a hand on the trophy, the horror of a QPR goal and another, followed by the slow, awful realisation that you’ve handed the cup to the arch-rivals (oh how we Protea fans know this all too well). Only to – miraculously – score two goals in three minutes to win it all. For the non-believers, that’s why we follow sport.

The scenes at Etihad Stadium reminded me of South Africa’s own recent sporting success stories. Although we won only one match, our hosting of the 2010 FIFA World Cup will probably remain one of my most vivid memories. It was an awesome event: two friends and I traveled around South Africa to watch eight live matches, including Bafana Bafana’s spectacular win at a jam-packed Bloemfontein stadium. It was a month of passion, of joy, of little productivity and lot’s of partying; in short, a month to be proudly South African. Or so it seemed to those of us attending.

I Can’t Get No Satisfaction: There is no evidence that the 2010 FIFA World Cup (the second and third dots from left) improved life satisfaction levels for South Africans

But our views are often shaped disproportionately by our immediate surroundings, which is why social scientists use surveys to assess the overall impact of a government policy, or to test the general perspective of South Africans. The National Income Dynamics Study is a national panel data set for South Africa, and recently released its second round of survey data. Although it is principally aimed at investigating poverty, income , unemployment and education trends,  it also asks some rather random questions that allow researchers to explore all sorts of interesting hypotheses. One of the questions asked in the survey, for example, was how satisfied the respondents are with their lives in general. (You may think this is pretty arbitrary, but there’s actually a growing literature about “happiness economics”. Bhutan famously measures their Gross National Happiness.) The inclusion of this question in the survey, and the fact that respondents were surveyed between May 2010 and April 2011 (the World Cup kicked-off in June 2010), allow us to see whether the World Cup had any effect on South Africans happiness. (A big thank you to Marisa Coetzee who helped with the analysis.)

But why would the World Cup affect life satisfaction? There is a rather large literature that suggests that mega-events are rather unprofitable ventures for countries: the returns, except in exceptional cases like the Sydney Olympics, rarely outweigh the costs. But countries continue to bid for these events, which must suggest that there are other, perhaps unquantifiable, benefits. Some scholars have suggested that these benefits include the “happiness”, “good image”, or “national pride” that these type of events create. Think of the “nation-building” following South Africa’s 1995 Rugby World Cup win, or the 2006 FIFA World Cup that portrayed Germany as a friendly, welcoming country.

Black and white: White South Africans show clear improvements in their life satisfaction measures during the 2010 FIFA World Cup

So do the NIDS survey show that mega-events (in this case, the 2010 FIFA World Cup) create large gains in terms of happiness? No.

The first graph shows the average life satisfaction across all respondents by month (the first month is May 2010, the 2010 FIFA World Cup started in June, i.e. dot number two). There is no evidence that South Africans were, on average, more satisfied with life in June and July than before or after the tournament. If anything, they were the least satisfied with life during those two months.

This picture changes remarkably if one considers only the white South African population. The second graph shows the marked improvement in white South Africans’ life satisfaction ratings during the months of June and July. Thereafter it stays relatively constant until December, when it plummets. (I don’t know why.) The basic problem here, though, is sample size. There are very few observations in the sample of white South Africans (not more than 20 for May), and so statistical significant deviations will be difficult to prove.

While there seems to have been at least some positive “happiness” effect of hosting the World Cup for white South Africans, there was no discernible impact for South Africans as a whole. Sport – and the passions it inspire – is seemingly a luxury good.

PS: Dieter von Fintel raised some issues with the graphs in the comments section. He requests to see 95% confidence intervals around the dots. Marisa again kindly helped to plot these, now with a continuous variable and a confidence band. The observations again begin in May 2010. Across all population groups, there is no evidence to suggest that during June or July life satisfaction increased.

Written by Johan Fourie

May 14, 2012 at 17:25

3 Responses

Subscribe to comments with RSS.

  1. Would love to see the confidence intervals around those scores. And understand the other circumstances that could have changed over that period (e.g. wage negotiations, job losses). Of course this “effect” is severely heterogenous and could perhaps be separated by socioeconomic groups, geography, etc. However, the sample may also only be representative of the 12 months as a whole, but not in each month separately, so that this type of analysis is infeasible. In May the enumerators perhaps happened to be in less affluent areas, reaching more affluent areas in June and July.

    dietervf

    May 14, 2012 at 18:17

    • Dieter – if all those things had been unproblematic, this would have been a paper rather than a blog post. I think we can say pretty confidently, though, that there was no significant increase in South Africans life satisfaction during the tournament, which is still quite a surprise.

      Johan Fourie

      May 14, 2012 at 18:34

      • I think if you had better data that was representative of any given population over each of the periods you may confidently be able to say quite the opposite. But we’ll never know, will we?!

        dietervf

        May 14, 2012 at 18:36


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: