Continuation of a previous musing

....I didn't finish a previous post on my own musings on trading, i.e. whether or not it makes any difference to the collective wealth of a nation. In retrospect, maybe it was a stupid question to ask - bad trades can surely decrease nations' wealth as is evidenced by the current recession, which is largely a consequence of sub-prime bets gone badly wrong.

So, it's not a "zero-sum game": while some people made money by going short on mortgage-backed securities, those who went long on the other side of the trade lost big time. This would classify as zero-sum on its own, but surely when looking at trading from a macroeconomic point of view, one cannot consider only the impact on party A and party B of a trade? In other words, the extent of systemic damage must be part of any trade-off.

But this then becomes a wider game-theoretic problem - in usual matrix-type choice-payoff games presented at game theory lectures, the "game" only considers payoffs to the players, i.e. the impact of their strategies on any one else outside the game is, by definition, excluded. Therefore, (Nash) equilibria in such games, while perfectly rational and optimising for the players, may not be so for people not directly participating in the game.

One way to incorporate this impact on the outside world is to link payoffs of a strategy to its impact on the outside world. A representative agent of the outside world, e.g. government or local regulator would have to enforce this link.

....More on this later.