Cries of "racism" have surrounded Labour's release of data on the impact of foreign buyers on the Auckland property market. But what's really upsetting people?

When are numbers racist?

That's a question I've been pondering for more than a week now, since I first saw Labour's data on people buying houses in Auckland, leaked to them by someone at the city's largest real estate firm (given it accounted for 45% of the sales in a three month period, I can I can safely assume there is none bigger). House sales information records name, address, price, date and, well, not much more. Yet that sparse data set has sparked huge debate since Saturday, because of Labour's choice to breakdown the information by surname.

The data was three months of house sales in Auckland by a single firm – from February to April this year. It totalled 3922 sales, of which 39.5 percent were estimated to have been to people with ethnically Chinese names. The census shows Chinese New Zealanders make up nine percent of Auckland's population, so the key question out of all this is where that other 30 percent of sales comes from.

Could it be immigrants or multiple house-buyers? That could certainly account for some... but how much? Or is it mostly buyers living offshore?

The real politik is that any politician who releases 'race-based' information can expect to generate controversy; it's kicking a political wasps' nest. Labour knew perfectly well that, however they spun the message that they are concerned about all foreign buyers, this would be viewed by many as an attack on Chinese.

Yet it's been fascinating to watch this story unfold. Having spent some time looking hard at the data, it's been interesting to hear ill-informed comments from people who have obviously barely looked past the headlines and see the story spin and distort away from where it began.

Such is politics. Sigh.

So is this data racist? Those such as Keith Ng are taking issue with the numbers themselves as unreliable. Which they are, up to a point. Ethnicity and residency are, of course, totally different. But then he's attacking a strawman when he makes that case because no-one has said otherwise.

My response to that is that it's good to debate and question the data (and Keith's made a couple of very salient points), but the fact is this is as useful a bunch of numbers of foreign buyers as we have in this country, given that nothing else is collected. Given that it's a topic of significant public interest, surely even this snapshot is useful to throw more light on the issue, even if it turns out not to be indicative of the bigger picture.

Reportage of many public issues is done with a piece of imperfect information here and a fragment of the puzzle there; over time a picture builds up.

I don't think it's racist to ask those questions about why there's such a big gap between the ethnic Chinese population and this snapshot of house buying. So are people, instead, really criticising Labour's choice to release the information?

Because some are just throwing the "racist" label around – saying Labour is "making it about race" or "picking on the Chinese" – without seemingly knowing exactly what they are criticising, what the numbers show or how they were processed.

[Update: The Maori Party has put out a statement this afternoon in which co-leader Marama Fox says, "These types of campaigns are exactly what Māori communities have been faced with for decades.  It is never ok to discriminate against any person because of their race”. What is she talking about? Campaigns? Who's discriminating by race? Does she even know what the data shows?]

So let's break this down by asking a few questions.

Was it racist when, in 2013, Tony Alexander from BNZ released his estimate of foreign buyers in the Auckland market? He reckoned about 11 percent of sales were to foreign buyers and broke it down further according to country.

Is it racist to acknowledge that China is a country with a vast population and a rapidly growing investor class which has only recently been allowed to start investing significant amounts of money overseas? And one where you can't own land yourself? Because there are solid reasons why your analysis of China and its citizens' interest in New Zealand property could be quite different from any other country.

Is it racist to acknowledge that it is easier for foreigners to buy houses in New Zealand than many other countries, given we have no land tax, no stamp duty, no requirement to 'build to buy', and only a little bit of a capital gains tax?

Surely not.

So the question comes down merely to the choice to break that data down by surname... and to release it. The surname breakdown is undoubtedly provocative and questionable. For me, it's useful information; still I understand why people say they can hear the dog whistle. But equally, the analysis of surnames is not as hit and miss as many have tried to make it out.

Many are lazily saying Lees, for example, come from all sorts of backgrounds and how ridiculous is it to assume they're Chinese. Except that's not how the calculations work. The surname part of the equation was based on probability.

As it was explained to me, Lees were weighted as 48 percent likely to be European, 40 percent Chinese, 0.3 percent Maori and 0.9 percent Other. By contrast Lis were 96 percent likely to be ethnically Chinese. Different again, Hoteres were 89 percent probably Maori. And so on. Chinese names weren't singled out; it was just that surprisingly high 39.5 percent figure that stood out and was thus focused on.

Those percentages were based on the electoral roll.

Someone may be able to tell me how that is statistically unsound, but it doesn't sound irrational to me. And it certainly wasn't just a matter of looking at Chinese-ish surnames and saying "that's probably one". And it wasn't about picking out those who look different or aren't 'one of us'.

As for the conclusion that someone, least alone a political party, would receive this leaked information and not release it, well, that seem farcical. Some seem to be implying that because the data is only a snapshot, Labour should have binned it.

Let's be real: No party would do that, and neither should they. While far from perfect, these numbers help build a picture of an issue of public interest. Lots of evidence of a problem or issue only comes by piecing together pieces of partial information so you can get a sense of the whole.

So what I'm left with is the impression that people seem to be more offended by the size of the key 39.5 percent number than anything else.

If you don't believe me, as yourself this: If the percentage of buyers had been 10 percent or 20 percent, would we have seen the same commotion? Would this have been labelled racist in the same way?

Further, if an official foreign buyers register one days finds the same, or a similar, result, would that be considered racist?

Answer that question and I think we get closer to the truth.

Some, it seems, just don't like the result or the tension this issue – and now this data – creates. I agree that Labour opened the door to racial stereotyping, but I also think some of the strongest critics of the release have been doing the most stereotyping.

A debate about people's ethnicity and involving nationalism comes with risks and has to be had carefully, but that's no reason not to have it.