In her article last week, Julie Fairey makes the claim that "research shows that for every poker machine, 0.8 of a problem gambler is created, which means with 230 new pokies we can estimate 184 new problem gamblers".
There are three issues with this statement: first, it is fallacious, secondly, the 0.8 figure is too high, and thirdly, it is mostly irrelevant in the context of problem gambling in Auckland.
The relevance issue is the most important. The SkyCity casino licence has been extended out to 2048. In that time Auckland's population is predicted to increase by a million people.
Given that 1 to 3 per cent of the adult population become problem gamblers this means 10,000 to 30,000 new problem gamblers.
The extra 184 created by the new poker machines is insignificant.
The statement is fallacious - it doesn't follow from the research.
It claims that poker machines create problem gamblers. That is not a fact - it is a hypothesis. Fairey is using the results from a 2009 paper "Access or adaptation?", a meta-analysis of surveys of problem gambling prevalence in Australia and New Zealand with respect to concentration of electronic gaming machines.
That paper found that, over the past two decades in Australia and New Zealand, as the number of poker machines per person rose the incidence of problem gambling rose with it and that the ratio was 0.80 people per machine.
That is evidence for the hypothesis but not proof.
The number of poker machines increased but other changes in society may have caused the increase in problem gambling. For example, there have been significant changes in gender roles, poverty, ageing population, unemployment, immigration, divorce rates, advertising, internet usage, lotto prizes and video game usage.
Any or all of these could have been responsible for the rise in problem gambling. You cannot credit it solely to the rise in poker machines.
Research* shows that poker machines are concentrated in the more deprived areas of New Zealand. This shows that the machine number can respond to market forces. It could be that every 0.8 new problem gambler creates enough demand for operators to install one new poker machine.
So what is the cause - do poker machines create problem gamblers or do problem gamblers create poker machines or do other factors create both? It is impossible to tell from the research. It is fallacious to imply that poker machines are singularly responsible for increases in problem gambling.
Finally, the 0.8 figure is plainly too high. It is an average over old and new poker machines and casino and non-casino locations and is too high for the specific case of adding 230 new machines to 1,649 existing machines in a casino.
For example, it ignores the law of diminishing returns. The first poker machine produces the most problem gamblers. As more machines are added the number of new gamblers per new machine drops until eventually adding new machines has no effect because all the problem gamblers are already hooked.
It ignores the location of the new machines. Studies suggest that non-casino machines are more conducive to problem gambling than casino machines. Also they are more visible to the community and so more likely to attract new gamblers.
Finally, the relationship between machines and problem gamblers seems to be weaker for casino poker machines than for non-casino machines^.
So the true number of new problem gamblers will be much less than 184 and may even be as low as zero.
*Pokies and poverty: problem gambling risk factor geography in New Zealand.
^An Exploratory Study of Problem Gambling on Casino Versus Non-casino Electronic Gaming Machines.
Tony Cooper is an Auckland quantitative analyst.