On Wednesday the Sydney Morning Herald reported the release of the 2010 Mercer annual quality of living survey with the headline, “Sydney beats Melbourne in world’s top cities league”.

This is not news. Sydney beat Melbourne in the 2009 Mercer survey too. Sydney has stayed in 10th position and Melbourne has “slipped” from 17th to 18th out of 221 cities across the world.

Victorian politicians prefer to reference the annual survey done by The Economist Intelligence Unit. Its 2010 Global Liveability Report ranks Melbourne 3rd after Vancouver and Vienna. Sydney is ranked 7th.

Do these surveys really indicate that Sydney is more “liveable” than Melbourne, or vice versa? No, they don’t.

For one thing, the difference in scores is miniscule. In the Mercer survey, Sydney scored 106 points to Melbourne’s 105. In The Economist’s survey Melbourne scored 97 and Sydney 96.

Clearly rankings give a misleading impression of the two cities relative merits.

These sorts of surveys have been criticised on a number of grounds, including lack of transparency about their methodologies, definitions and quality of data. But that criticism misses the point that they are designed for a different purpose – to assist companies determine living allowances for staff posted to an overseas destination. The lower the city ranks, the higher the compensating allowance.

Hence they are not good at measuring a city’s liveability from a permanent resident’s point of view. For one thing, expatriates are usually executives assigned temporarily to an overseas posting. They can overlook shortcomings that might worry a permanent resident. They are likely to be highly paid, to enjoy an expense account and to live in the best and most salubrious part of any city they’re posted to.

I’ve not lived in either Hawthorn or Reservoir, but I imagine they’re pretty different – and I wouldn’t expect to see many overseas executive assigned a house in Reservoir. Expats are also likely to be well educated with the tastes and wherewithal to enjoy the sorts of attractions, like opera, that the average permanent resident either has little interest in or cannot afford.

For her part, the average Melburnian is more likely to be concerned with matters like job opportunities and housing affordability, neither of which is of interest to the expatriate and neither of which is addressed directly by the two surveys. She is also likely to live in the middle or outer suburbs – possibly in Reservoir because houses sell there at the Melbourne median price – and to be more concerned about the quality of State schools than private or international schools.

These sorts of surveys are also of limited value in making intra-country comparisons. They give a high weighting to variables that reflect national characteristics rather than city-specific characteristics, so their ability to discriminate between places in the same country isn’t high.

All Australia’s capitals tend to rank highly simply because Australia ranks well on a host of essentially national characteristics like personal safety, education and health services. For example, the Mercer survey has Sydney on 106 points and Brisbane, the lowest ranking Australian city after Melbourne, Perth, Canberra and Adelaide, on 102 points.

So these surveys aren’t really very enlightening about “liveability” at the city level. We should stop believing our own bullshit about Melbourne’s liveability until some hard evidence is placed on the table.

But what, if anything, makes one city more “liveable” for permanent residents than another? I’ll have a look at that question in part 2, next week.