The Runaway Bailout

That was January. This is now.

Some figures in yesterday’s Globe and Mail regarding the bailout of GM and Chrysler, and in particular the support of the pension funds, made for pretty shocking reading:

Ottawa and Toronto were already asking a lot of Canadians – most of whom have no private retirement fund and earn significantly less than auto assembly workers – by allowing some of the bailout money to go toward fixing an estimated $7-billion shortfall in GM Canada’s pension plan.

But with the latest forecast pegging the overall bailout bill at as much as $13.5-billion, or more than three times the original estimate, politicians are testing the limits of recession-racked Canadians’ tolerance and financial wherewithal. The ballooning bailouts are pushing Ottawa deeper into the red, with this year’s deficit projected to surpass $50-billion.

At General Motors of Canada Ltd. alone, the rescue package could amount to a staggering $1.4-million for every job saved, with no guarantee that the bailout will ensure the long-term survival of the company’s remaining auto assembly and engine plants.

I don’t have the latest figures on what GM workers earn, but two years ago the average gross hourly wage in the US was US$39.68 (£24.60 at current exchange rates), which becomes $73.26/hour (£45.42) when benefits are taken into account. In comparison, this latter figure is about $30 more per hour than the equivalent rates paid in US-based Japanese plants. The auto workers unions have made concessions as part of the negotiations with GM and Chrysler to bring their wages closer to those in Japanese plants, but the question remains as to how the average tax payer, who is likely to be receiving very little support in their industry of employment and to be earning considerably less than the wage rates mentioned above, is left paying $1.4 million per worker to the pension funds of mismanaged (and grossly overpaying) companies in one particular sector.

Unfortunately, the Government of Ontario doesn’t come out of this affair in a particularly good light. In 1992 Ontario decided to exempt companies with over CAN$500 million in assets from its plans to force all companies to fully fund their pension schemes, under the assumption that such companies were ‘too big to fail’. The Canadian Auto Workers union is now arguing that the Ontario Government was negligent, and is on the hook for the shortfall. Although the government has claimed not to have the money to fund the amount, recent reports suggest that either the provincial and/or the national government are likely to end up footing the bill.

The automakers’ lobbyists also haven’t been slow to take advantage of the industry’s geographical spread across the US-Canadian border. If one government played ball and the other one didn’t, you could expect a massive migration of automotive supply chain across the frontier. In the end, Stephen Harper, not a man with a great history of supporting government intervention in the private sector, felt he had no choice but to hold his nose and match Obama’s guarantees or see vast swathes of southern Ontario jobs disappear into thin air. Since that time, the bailout sum has ballooned to potentially three times the amount originally anticipated.

Canadians, and particularly those living in Ontario, will be paying for the bailouts in tax rises and government cutbacks for many years to come, and given the automakers’ previous reneging on promises, there is little guarantee that the money will see a long term future for GM and Chrysler in Canada. GM’s workforce in Ontario has fallen from 20,000 five years ago to 14,000 in advance of the recent difficulties, and a planned 7,000 by next year, not taking into account any further cut backs deemed necessary in the forthcoming bankruptcy proceedings. The argument goes, however, that there are more than just the jobs at GM and Chrysler at stake. Both companies would be likely to bring down much of their supply chains with them if they disappeared or relocated, and as a result would endanger production at the other car manufacturing plants in Ontario.

Surely, however, the costs of supporting the affected automotive suppliers in order to guarantee continued supply to Ford, Honda and Toyota, and of retraining and sensibly compensating those thrown out of work by the exodus/collapse of GM and Chrysler would be considerably lower than the mind-boggling sums now under consideration and, perhaps more importantly in the long run, would avoid the dangerous precedent set by such lavish rewarding of failure and excess.

Hedgehogs and foxes at The Guardian

Two opinion pieces in the Guardian this week on the subject of swine flu influenza A (H1N1) provided an interesting example of Philip Tetlock’s classification of good and bad pundits as ‘foxes’ and ‘hedgehogs’, as I outlined in another recent post.

On Wednesday, Simon Jenkins published a column entitled “Swine Flu? A panic stoked in order to posture and spend”. Here are some extracts:-

Health scares are like terrorist ones. Someone somewhere has an interest in it. We depend on others with specialist knowledge to advise and warn us and assume they offer advice on a dispassionate basis, using their expertise to assess danger and communicating it in measured English. Words such as possibly, potentially, could or might should be avoided. They are unspecific qualifiers and open to exaggeration.

The World Health Organisation, always eager to push itself into the spotlight, loves to talk of the world being “ready” for a flu pandemic, apparently on the grounds that none has occurred for some time. There is no obvious justification for this scaremongering. I suppose the world is “ready” for another atomic explosion or another 9/11.

[...]

During the BSE scare of 1995-7, grown men with medical degrees predicted doom, terrifying ministers into mad politician disease. The scientists’ hysteria, that BSE “has the potential to infect up to 10 million Britons”, led to tens of thousands of cattle being fed into power stations and £5bn spent on farmers’ compensation.[...] This science-based insanity was repeated during the Sars outbreak of 2003, asserted by Dr Patrick Dixon, formerly of the London Business School, to have “a 25% chance of killing tens of millions”. The press duly headlined a plague “worse than Aids”. Not one Briton died.

The same lunacy occurred in 2006 with avian flu, erupting after a scientist named John Oxford declared that “it will be the first pandemic of the 21st century”. The WHO issued a statement that “one in four Britons could die”.

Simon Jenkins is an arch-cynic. This often serves him well, making his columns iconoclastic and insightful. Having cynicism as your default setting, however, is sometimes going to result in wrong-headed opinions such as those in this piece.

His insistence that health specialists should avoid using words like ‘possibly, potentially, should and might’ clearly indicates his pedigree as a columnist. While pundits may use the illusion of certainty to talk up the importance of their opinions and predictions, it would be utterly irresponsible of the WHO and other medical experts to speak in terms of certainty about a newly discovered virus with an inherently chaotic infection pattern. Also note that later in the article Jenkins criticises John Oxford for stating that avian flu ‘will be’ the first pandemic of the 21st century. Would Jenkins have preferred that he hedged his bets with a ‘potentially’ or a ‘might’? It seems difficult to win in Simon Jenkins’ eyes.

He then criticises the WHO for stating that the world is ‘ready’ for a flu pandemic, and for scaremongering. I presume what he is referring to is the WHO’s Keiji Fukuda’s comments on Monday:

“I believe that the world is much, much better prepared than we have ever been for dealing with this kind of situation[...] The past five years have put us in (the) best possible position to handle this kind of situation.”

Jenkins can’t have it both ways. If the WHO are stating that the world is ready, surely that is reassuring, not scaremongering. When he goes on to mention the BSE, SARS and avian flu scares, he is talking as if scientists at the time had the perfect knowledge to predict that the diseases wouldn’t spread. At least in the last two instances, without extremely risk averse containment procedures, things could have been a lot worse. The millions of deaths in the 1918 pandemic are surely sufficient reason to act with extreme caution and present people with the pandemic scenario in order that they take seriously the necessary simple steps (frequent washing of hands, for example) that can impede the spread of the virus.

On Wednesday, common sense returned to the Guardian’s pages in a piece by Ben Goldacre, whose ‘Bad Science‘ column and website have tagged him as an arch-enemy of overhyped scientific claims:-

Just like with Sars, and bird flu, and MMR, is this all hype? The answer is no, but more interesting is this: for so many people, their very first assumption on the story is that the media are lying. It is the story of the boy who cried wolf.

We are poorly equipped to think around issues involving risk, and infectious diseases epidemiology is a tricky business: the error margins on the models are wide, and it’s extremely hard to make clear predictions.
[...]
All people have done is raise the possibility of things really kicking off, and they are right to do so, but we don’t have brilliantly accurate information. Someone has said that up to 40% of the world could be infected. Is that scaremongering? Well it’s high, and I’m sure it’s a bit of a guess, but maybe up to 40% could be. Annoying, isn’t it, not to know.

Someone has said 120 million could die. Well I suppose they could: I’m sure it was done on the back of an envelope, by guessing how many would be infected, and what proportion would die, but I don’t think anyone’s pretending otherwise.
[...]
By Tuesday, pundit-seekers from the media were suddenly contacting me, a massive nobody, to say that swine flu is all nonsense and hype, like some kind of blind, automated naysaying device. “Will you come and talk about the media overhyping swine flu?” asked Case Notes on Radio 4. No. “We need someone to say it’s all been overhyped,” said BBC Wales.

[...] Simon Jenkins suggested the same thing. It’s not true, I said. They were risks, risks that didn’t materialise, but they were still risks. That’s what a risk is. I’ve never been hit by a car, but it’s not idiotic to think about it. Simon Jenkins won’t be right if nobody dies, he’ll be lucky, like the rest of us.

Goldacre and Jenkins are both known for calling out BS in their columns, but Goldacre isn’t defined by his skepticism. He can easily slip from his default mode to a robust defence of the coverage of the potential pandemic. While Jenkins appears the one-dimensional cynical hedgehog, Goldacre proves himself to be the multi-faceted fox.

It is, of course, very possible that this won’t be as bad as is feared, but as long as it can’t be ruled out, it makes a lot of sense for people and institutions to take practical steps to be ready just in case. This isn’t a bogus threat or conspiracy theory. This is a regular punctuation mark throughout the course of human history.

Although much is uncertain at the moment, the breakout of the disease in Mexico means that looking at the situation down there gives at least some of idea of how things could be elsewhere in the world very soon if the virus continues to spread unabated. Unfortunately, the statistics coming out of Mexico are very unclear. The difference between confirmed and suspected deaths is large, and even the confirmed figure seems to go down as well as up. I personally suspect that the number of actual cases in Mexico is much higher. We may only seeing the tip of the iceberg – the extreme cases which have been reported. As we know, the symptoms in those returning from Mexico have been mild. If they had contracted the virus at home, how many would have gone to the doctor, and how many of those would have been tested? I imagine that many minor cases in Mexico have gone unreported for similar reasons.

This would potentially explain two anomalies: firstly, the high death rate in Mexico. If the infection rate was much higher than reported, this would bring the death rate down to a much more statistically understandable level, meaning that many more cases would be needed outside Mexico before a proper comparison could be made. Secondly, it would make more sense of the rate of infection among those returning from Mexico compared with the rate within the country. If I’m right, this would suggest a high infection rate but with a much lower mortality rate than currently reported in Mexico. Of course, I’m no expert and this is just a homespun theory. I’m just trying to make sense of some perplexing inconsistencies in the stats. I’d be interested to know what others think.