What we really need are statistics that show the age of death of those people who survived until an arbitrary retirement age - eg 65. That is the only statistic that would really support the claim "people are living longer" in a way that actually means what it appears to be trying to convey. Ie, is "dying of old-age-related events" happening at an older age than before.

For those with a specific interest in the North West, I have just received this:

http://www.nwriu.co.uk/r...s/population_change.aspx

REFUSE COLLECTORS PAYING FOR BANKERS' PENSION!
Living longer so increase the pension age? Well another set of statistics says that manual workers have a worse life expectancy than sedentary workers. Another says the well off have a better life expectancy than the poor. In fact manual workers life expectancy has increased only marginally over recent decades. So as they die, on average, shortly after retirement their contributions go to help us pay the long living bankers state pension.
Two last questions - Who has done more for us in the last decade, the bankers or the refuse collectors? Who deserves the £7 billion bonus this year (or will it only be £4 billion)?

Foggiest,

it's a well known fact that refuse collectors contribute more to public health and hence save more lives than surgeons.

Peter Jason Taylor

Of course projected life expectancy figures are an estimate. But the actuaries and statisticians making these estimates are very much aware of this and readily agree with you. They are highly skilled people making their best judgements. If a doctor says you have an estimated six months to live, do you say that a new drug might be found tomorrow and instantly cure you? Maybe you would but still know that the professional is likely to be right under all the known circumstances.

Would you prefer to muddle along with no forecasts at all? In the days when life assurance first began that is precisely the way business was conducted. The results were that some people were ludicrously overcharged and others the reverse. Insurers went bust and survivors went unpaid. The story is told in an excellent book "Against the Gods: the remarkable story of risk" by Peter L Bernstein.

Dennis

Similarly the manufacturers on machinery cannot run tests for 100,000 hours before launching a new product. I am not sufficiently well versed in their statistical methods to tell you how, but I believe their estimates are generally sound and based on knowledge of the materials, the normal operating environment and experience of making similar estimates in the past. It is how engineers earn their living.

Plants to manufacture this kind of computer component require enormous investment. What manufacturer is going to make that investment and destroy their reputation and future sales by selling shoddy products and making egregiously false claims? Maybe you can come up with a few examples but on the whole I think they will be exceptions that prove the rule.