Actuaries working in private companies where they're working with their own money (and constrained by the profit motive) are always going to come up with risk assessments that are biased away from anything that might cost them money and biased towards anything that's beneficial (during the financial outlook period) to their organisation.
Governmental actuaries, working in an arm's length environment where they're free to use whatever models and methods they like (and aren't going to get hit if they come out with bad news) will probably produce a much more realistic risk outlook.
Businesses which are large enough to retain their own staff of actuaries are unlikely to go out of business because of a bad projection. That's what that odious (and becoming ubiquitous) "Past performance is no guarantee of future results, and the Company accepts no liability blah blah blah" boilerplate is about.
Don't say "spreads the risk onto others" like it's a bad thing. Risk pools are actually your friends...
That said, I've been the victim of some badly designed government software. The software in question was developed by a team of consultants hired by Andersen Consulting, who were in turn contracted with the government. If the government had done it in-house, probably it would have cost 1/10 as much (for how much can you directly hire a programmer or ten in the Province of Ontario -- a fuck of a lot less than you can hire a $300/hr consultant!) and actually worked when it was done. It would have also put Ontarians to work, instead of US consultants. Everyone wins.
(no subject)
Governmental actuaries, working in an arm's length environment where they're free to use whatever models and methods they like (and aren't going to get hit if they come out with bad news) will probably produce a much more realistic risk outlook.
Businesses which are large enough to retain their own staff of actuaries are unlikely to go out of business because of a bad projection. That's what that odious (and becoming ubiquitous) "Past performance is no guarantee of future results, and the Company accepts no liability blah blah blah" boilerplate is about.
Don't say "spreads the risk onto others" like it's a bad thing. Risk pools are actually your friends...
That said, I've been the victim of some badly designed government software. The software in question was developed by a team of consultants hired by Andersen Consulting, who were in turn contracted with the government. If the government had done it in-house, probably it would have cost 1/10 as much (for how much can you directly hire a programmer or ten in the Province of Ontario -- a fuck of a lot less than you can hire a $300/hr consultant!) and actually worked when it was done. It would have also put Ontarians to work, instead of US consultants. Everyone wins.