$139,000?

Ah, the lazy summer season, when lobby groups can sneak pretty much any kind of rubbish into Radio New Zealand's Inbox and have it be treated as news. Here's the latest:

Cigarettes rose in price on 1 January, for the third time in 21 months.

The price of a packet of 20 cigarettes is now about $14 and a pouch of loose tobacco is about $31.

The Quit Group estimates each smoker costs the economy $139,000 per year in health and other costs.

Chief executive Paula Snowden says the group returns $38 to the economy for every $1 it is funded, but has had no increase since 2007.

She expects the price increase to save the economy up to $73 million.

Let's start with what's right about the piece. I have no reason to believe that they've lied about the price of a packet of cigarettes. The rest isn't so good.

Recall that GDP per capita in New Zealand is somewhere around $45,000*. So Snowden is claiming that a smoker costs "the economy" three times per capita GDP. And recall that median income is less than per capita GDP, which also includes payments to capital. Snowden's number is three times that. It's utterly implausible.

Next, she expects the price increase to save the economy up to $73 million. If flipping a smoker from smoking to non-smoking saves those costs, then she's expecting about 525 people to quit. If only a fifth of costs are avoidable and the rest are sunk, the price increase results in around 2600 people quitting smoking. But the last survey numbers I'd seen have about 22% of the population aged 16-65 smoking. Again, see asterisk below, but there are about 4.5 million people in the country. Let's be conservative and say 300,000 smokers; 2,600 quit. This is all ballpark, but it suggests maybe one percent of smokers are expected to quit with the price hike. And, that's with me granting the $73 million. I'd really need to see the workings to figure out what they're doing here.

Finally, let's aggregate up for a final plausibility check. Say there are 300,000 smokers in the country. If they each cost $139k, that's about $41 billion that Paula Snowden claims smokers are costing the country. And even the ridiculously high Collins & Lapsley numbers have smoking costing all of Australia only $32 billion. And there are just a few more people in Australia than in New Zealand.

Every decent journalist should have a few of these numbers just rattling around in the back of his head to give numbers like Snowden's a rough plausibility check. It's not plausible that a smoker costs the country three times per capita GDP. That's the first bit of stink. If the implied participation elasticities are as suggested by a quick ballparking, then we'd wonder about the costs imposed on poor people who continue to smoke even if everything else is right. That's the second bit of stink that should have been queried.

But the biggest howler is that Snowden claims a 38:1 benefit to cost ratio for her organization's work.

When did they stop teaching scepticism in journalism school? Radio NZ, you've been pwned by The Quit Group. You can do better.

*Please note that I'm working off memory for a lot of these figures. No Internet at the guest house in Nelson where we've been avoiding earthquakes for the last week, so I'm burning through Vodafone Pay-As-You-Go data. Back in quakeland tomorrow.

Previously: Bogus MoH figures on smoking costs.

Update: Sarah Woods of The Quit Group advises that the $139,000 figure is the present discounted value of costs, with the estimate based on Des O'Dea's prior figure (which included a raft of costs borne by smokers that don't really count as costs to "the economy") and with ROI calculations undertaken by BERL. I'll have a look through their report. But Radio NZ has inflated The Quit Group's figure by an order of magnitude by transforming a present discounted value figure into an annual cost figure. At an 8% discount rate, the PDV of $139k is about $1.7 million. And let's remember exactly what the O'Dea report concluded about the costs smokers impose on "the economy":

Leaving aside these difficulties, it is certainly reasonable to assume that most of the additional health-care costs caused by smoking are borne by non-smokers through additional taxes (smokers do pay some share of these taxes). Also it is reasonable to assume that most of the 'lost  production' costs of premature mortality and increased morbidity are borne by smokers and their households (though there is some loss of profits also, and of tax revenue to government). A considerable amount of work would, however, be needed to get precision on these matters.

Without trying to calculate a precise estimate of 'external costs' it does seem reasonably apparent that the tax contribution of approximately $1 billion annually by smokers exceeds substantially the external costs of smoking which fall on non-smokers. If savings on pension costs from premature mortality of smokers were added as well the net fiscal contribution of smokers, to the fiscal gain of non-smokers, would be further increased. [emphasis added]