Go for Olympic host city gold - and go for broke too

So he has been on both sides of the Brisbane deal in a situation it is hard to see occurring anywhere else in politics or commerce, and gives an idea of how the IOC conducts business. Coates excusing himself at times to avoid conflicts of interest seems enough to keep everyone happy. And it helps that Brisbane was the only bidder anyway.

To be fair to the IOC, it is aware of the expense of the Olympics and is planning to reduce costs under its "Olympic Agenda 2020". However, it still takes none of the financial pain, and where it has trialled some of these strategies – Rio, the 2018 PyeongChang Winter Olympics, and Tokyo – the budget overruns have still never been less than 100 per cent.

Brisbane's budget for the Olympics is $5 billion. This would make it the cheapest since the Athens Olympics, and even cheaper than the 2000 Sydney Olympics, which sounds ambitious. According to the Flyvbjerg study, the amount has at least a 20 per cent risk of being $15 billion or higher.

Brisbane is in a better position than most. Three years ago it staged the Commonwealth Games, and according to modelling done for the Queensland government by Griffith University, the Games cost $1.6 billion, or 26 per cent over budget, a creditable result.

Yet this Queensland government is showing poor project management skills, with the $5.4 billion Cross River Rail blowing out to $6.9 billion (disguised through a sneaky privatisation of the stations on the route), and another $1 billion may now be added to that because of changes to the track and station locations. That's a 46 per cent cost overrun.

And the state government has also done a sweetheart deal with the CFMEU for government construction, meaning that a carpenter could earn $254,788 for a 50-hour week, or $74,688 more than a government backbencher. That's bound to make the eye-wateringly expensive $1 billion redevelopment of the Gabba to provide another 4000 seats and a warm-up athletics track even more expensive.

What is the upside?

Typically, governments claim employment and economic benefits, and modelling is conducted to prove these claims, but can we find any real-world statistics to justify projections?

The Commonwealth Games has to be the best gauge for Queensland's likely Olympics performance, and the Griffith University study claims a modelled increase of 1.266 million visitors.

The real measure of this should be overseas travellers, otherwise it is just an exercise in stealing domestic tourists from other states.

In 2018 there was an Australia-wide increase of only 5 per cent in short-term arrivals, the worst year since 2013. The legacy effects were non-existent, with only 2 per cent the year after, and no point bothering with 2020 and 2021.

Griffith also forecast an increase of $2.5 billion in gross state product over the period 2012-22.