![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Accountancy as faith-healing
Jul. 31st, 2008 11:22 amPrudential UK's first-half figures this morning raised an interesting accountancy question. Casually posting a £1.95bn writedown, Prudential referred to these as "short-term fluctuations in investment returns", to which my immediate response was "how do you know?"
I think it's sad that accountants/auditors will sign off stuff like this. In a way, I suppose, it's like the "fair value" argument (marvellously recounted at a micro-level by the line in one of the property shows that "this house is worth £400k, but no-one will offer you that at the moment") which, by definition, requires accountants to be soothsayers.
I've just begun reading the Baroque Cycle by Neal Stephenson (and it's a cracker -- what a find!) and it reminded me how little had changed in the world of "science", with mystery and hokum wrapped up in what look like objective numbers. In the current situation, three systems of accounting (IFRS, EEV, GAAP) can give three totally different figures for the year for a life assurer. It's also quite possible for a company to release perfectly sound figures, quite correctly signed off by the auditors, and then to disover a week later that it is insolvent. By any definition, this means that today's accounting systems are flawed. If an accounting system cannot show a roughly accurate state of affairs in a company, then it is an unsatisfactory system.
To refer to any change in value as "short-term" is nonsense. It's a bit like the definition of a long-term investment -- a short-term investment that went wrong. To even involke the concept of "fair value" is stupid.
In the old days, British Airways could fairly have been described as a pension scheme that happened to fly airplanes. In the US, Ford and General Motors might reasonably be called pension and health schemes that make a few cars. In volatile markets, companies with large levels of assets compared to their ongoing business ( a perennial problem for life assurers) see their wealth shoot up and down while income remains relatively constant.
Hotels used to suffer this problem when they held the freeholds to very valuable real estate. It was solved quite easily. Hotels are service industries, not property owners. By selling the properties and leasing them back, they eased their capital requirements enormously and got out of something they knew knothing about, focusing on what they did know about.
Life insurance companies need to decide what it is they actually do. If they insure lives, then they should outsource their investment work to other companies. That would ease capitalisation requirements and it would certainly simplify their accounting procedures. This has already been done with closed life books (see Resolution). Why not do it for open ones too?
________________
The second half of the year in poker land has, like the first half of the year, begun well. However, I continue to ponder possible new strategies and how to stay ahead of the game. One interesting number I noted among some players who claim to be winners (and I have no reason to disbelieve them apart from the fact that I tend to disbelieve most poker players) was a won-money-at-showdown percentage of about 47%. I'd been clicking along at 52% for a while and had decided that this was too low. I've pushed it higher (mainly by making more folds pre-flop), and I'm winning more, but that doesn't mean that the lower winning percentage at showdown is wrong.
The clear conclusion (or so it seemed to me) was that these guys were pushing further if they were not being reraised. There are some hands where, if I have been called on the flop by a tight player, I tend to give up, calling or folding on the river depending on the marginal strength of my hand and on the size of opponent's bet.
These players with the lower showdown percentage must be pushing on to the turn and the river. They control the pot size by putting in bets that are large enough to look like standard progressions, but are smaller than when you are looking for a stack-off. Opponents need to be well-analysed. And it seems to me that if these guys have any hand at all, they are prepared to keep betting on turn and river, because the folds that they elicit en route more than make up for the increased number of showdowns that they lose at the end.
The important thing is not to push this theory too far. We are only talking about a maximum of 10% of showdowns (47% wins compared with 57% wins).
I suspect that the argument of "worse hands won't call and better hands won't fold" is overstated. Other people do strange things, and I reckon that this sustained-but-controlled aggression on turn and river might generate folds that are unexpected. They might also generate calls from marginal losers.
I defended against a guy whom I thought was playing this way last night and I won (just) with a pair of nines Queen kicker against his pair of nines ten kicker. He bet out from the blind on the flop, turn and river (the last being particularly astute, I feel -- my hand by then was begining to look eminently foldworthy against any bet on the river that wasn't a bluff), but the bets were only just above half-pot. He was giving himself two ways to win, and you needed to be paying attention to see that the size of his bets compared with the size of our stacks made it look like a guy looking to keep the pot small. And he nearly got me to fold.
I like this approach ("aggressive betting for value?") and I think that it could win me a number of pots on the turn and river (and these are bigger pots, remember), even if it also increases the number that I lose at showdown. The trick is to make the sum of the first two outweigh the latter. I think that it would also suffer in games that were innately aggressive.
___________
I think it's sad that accountants/auditors will sign off stuff like this. In a way, I suppose, it's like the "fair value" argument (marvellously recounted at a micro-level by the line in one of the property shows that "this house is worth £400k, but no-one will offer you that at the moment") which, by definition, requires accountants to be soothsayers.
I've just begun reading the Baroque Cycle by Neal Stephenson (and it's a cracker -- what a find!) and it reminded me how little had changed in the world of "science", with mystery and hokum wrapped up in what look like objective numbers. In the current situation, three systems of accounting (IFRS, EEV, GAAP) can give three totally different figures for the year for a life assurer. It's also quite possible for a company to release perfectly sound figures, quite correctly signed off by the auditors, and then to disover a week later that it is insolvent. By any definition, this means that today's accounting systems are flawed. If an accounting system cannot show a roughly accurate state of affairs in a company, then it is an unsatisfactory system.
To refer to any change in value as "short-term" is nonsense. It's a bit like the definition of a long-term investment -- a short-term investment that went wrong. To even involke the concept of "fair value" is stupid.
"Ahh, yes, this was worth £600m last year, but at the moment we could only sell it for 10 pence. However, we think that the "real" value is nearer £400m, so that's how we will value it".
"Well, no, when someone is willing to pay you £400m, then you can value it at that. Until then, value it at 10 pence".
"But we can't do that! That would lead to ridiculous voltility in our results! It would overwhelm our underlying businesses."
"Tough".
In the old days, British Airways could fairly have been described as a pension scheme that happened to fly airplanes. In the US, Ford and General Motors might reasonably be called pension and health schemes that make a few cars. In volatile markets, companies with large levels of assets compared to their ongoing business ( a perennial problem for life assurers) see their wealth shoot up and down while income remains relatively constant.
Hotels used to suffer this problem when they held the freeholds to very valuable real estate. It was solved quite easily. Hotels are service industries, not property owners. By selling the properties and leasing them back, they eased their capital requirements enormously and got out of something they knew knothing about, focusing on what they did know about.
Life insurance companies need to decide what it is they actually do. If they insure lives, then they should outsource their investment work to other companies. That would ease capitalisation requirements and it would certainly simplify their accounting procedures. This has already been done with closed life books (see Resolution). Why not do it for open ones too?
________________
The second half of the year in poker land has, like the first half of the year, begun well. However, I continue to ponder possible new strategies and how to stay ahead of the game. One interesting number I noted among some players who claim to be winners (and I have no reason to disbelieve them apart from the fact that I tend to disbelieve most poker players) was a won-money-at-showdown percentage of about 47%. I'd been clicking along at 52% for a while and had decided that this was too low. I've pushed it higher (mainly by making more folds pre-flop), and I'm winning more, but that doesn't mean that the lower winning percentage at showdown is wrong.
The clear conclusion (or so it seemed to me) was that these guys were pushing further if they were not being reraised. There are some hands where, if I have been called on the flop by a tight player, I tend to give up, calling or folding on the river depending on the marginal strength of my hand and on the size of opponent's bet.
These players with the lower showdown percentage must be pushing on to the turn and the river. They control the pot size by putting in bets that are large enough to look like standard progressions, but are smaller than when you are looking for a stack-off. Opponents need to be well-analysed. And it seems to me that if these guys have any hand at all, they are prepared to keep betting on turn and river, because the folds that they elicit en route more than make up for the increased number of showdowns that they lose at the end.
The important thing is not to push this theory too far. We are only talking about a maximum of 10% of showdowns (47% wins compared with 57% wins).
I suspect that the argument of "worse hands won't call and better hands won't fold" is overstated. Other people do strange things, and I reckon that this sustained-but-controlled aggression on turn and river might generate folds that are unexpected. They might also generate calls from marginal losers.
I defended against a guy whom I thought was playing this way last night and I won (just) with a pair of nines Queen kicker against his pair of nines ten kicker. He bet out from the blind on the flop, turn and river (the last being particularly astute, I feel -- my hand by then was begining to look eminently foldworthy against any bet on the river that wasn't a bluff), but the bets were only just above half-pot. He was giving himself two ways to win, and you needed to be paying attention to see that the size of his bets compared with the size of our stacks made it look like a guy looking to keep the pot small. And he nearly got me to fold.
I like this approach ("aggressive betting for value?") and I think that it could win me a number of pots on the turn and river (and these are bigger pots, remember), even if it also increases the number that I lose at showdown. The trick is to make the sum of the first two outweigh the latter. I think that it would also suffer in games that were innately aggressive.
___________
