The fallacy of social economics - a rant
Basically the largest fallacy when trying to apply economic theory to social situations is the failure to reaize that economic theory is a synthetic statistical aggregate estimation of an expected pool of results. In other words the result of an economic prediction is only true within the context of the parameters of the prediction (within a larger group of outcomes), and those hardly ever can adequately mirror the parameters of reality. Truth is stranger (not to mention more diverse) than fiction.
To try to apply this to an individual POV is dangerous at best, and just plain insane at worst. Why? Because a statistical aggregate estimate has no bearing on individual outcomes. What do I mean by that? Okay, you're standing in front of a wall with a bunch of (say, 19) other people. Part of that wall is about to fall down and crush someone, but you don't know which part. But you do know that you only have a 5 percent chance of having it fall on you, so you don't feel that worried. Of course the other 19 people all feel the same way too so someones in for a horrible shock.
Of course the good economist could distribute the risk over the entire pool so that instead of one person being 100 percent dead, everyone ends up 5 percent dead. While mathematically just as valid, it still doesn't bring the dead person back to life. Mistakes like this happen all the time. Sometimes they're not mistakes. In fact the way we're taught as consumers to interpret statistics reinforces this skewing (and our sense of entitlement). We're always the first to acknowledge that the wall will fall, but always horrified when it falls on us.
Take it up another notch to public policy. Too many pizza boxes are being discarded by the road, so in order to cut down on litter the state passes a law that all pizza places must print their phone number on the pizza boxes and any pizza boxes found will result in a fine to the pizza shop. The theory is that this will provide and incentive to the pizza shops to find a way to reduce the littering of their customers (since the individual customers can't be apprehended). Of course the result on the individual pizza place may be
1.) We already have a high profit margin, so fuggedaboutit!
2.) We'll pass the cost of the fines along to the customer depending on what the fines turn out to be.
3.) We'll pre-emptively raise our prices to anticipate the fines, and if they don't occur it'll be extra profit.
And whichever result may depend on a lot of different factors like the competition and proximity of certain pizza place to/with each other. Owner's proclivity to raise prices. Individual shop profit margins, or a myriad of other individual factors. Naturally you can't incorperate *all* possible factors into an economic model, so there are apt to be holes in your predictions. Ah! Says the economist, that's not all that important as long as we hit the main causal factors because we're looking for an aggregate prediction anyway....and he's right. Except that many of these aggregate predictions have no bearing on the individual outcomes...good or bad. Not only that, it might not even be a good predictor due to localized variation (there's a No-Pizza ordinance in XYZ county, therefore no pizza shops). Yet, the prediction, because it is true in aggregate, may be used to support further public policy.
So beware of using large population predictions to justify (or predict) individual experience. And be wary of the same thing being handed to you by the media/government/corperate complex.
End of rant.
To try to apply this to an individual POV is dangerous at best, and just plain insane at worst. Why? Because a statistical aggregate estimate has no bearing on individual outcomes. What do I mean by that? Okay, you're standing in front of a wall with a bunch of (say, 19) other people. Part of that wall is about to fall down and crush someone, but you don't know which part. But you do know that you only have a 5 percent chance of having it fall on you, so you don't feel that worried. Of course the other 19 people all feel the same way too so someones in for a horrible shock.
Of course the good economist could distribute the risk over the entire pool so that instead of one person being 100 percent dead, everyone ends up 5 percent dead. While mathematically just as valid, it still doesn't bring the dead person back to life. Mistakes like this happen all the time. Sometimes they're not mistakes. In fact the way we're taught as consumers to interpret statistics reinforces this skewing (and our sense of entitlement). We're always the first to acknowledge that the wall will fall, but always horrified when it falls on us.
Take it up another notch to public policy. Too many pizza boxes are being discarded by the road, so in order to cut down on litter the state passes a law that all pizza places must print their phone number on the pizza boxes and any pizza boxes found will result in a fine to the pizza shop. The theory is that this will provide and incentive to the pizza shops to find a way to reduce the littering of their customers (since the individual customers can't be apprehended). Of course the result on the individual pizza place may be
1.) We already have a high profit margin, so fuggedaboutit!
2.) We'll pass the cost of the fines along to the customer depending on what the fines turn out to be.
3.) We'll pre-emptively raise our prices to anticipate the fines, and if they don't occur it'll be extra profit.
And whichever result may depend on a lot of different factors like the competition and proximity of certain pizza place to/with each other. Owner's proclivity to raise prices. Individual shop profit margins, or a myriad of other individual factors. Naturally you can't incorperate *all* possible factors into an economic model, so there are apt to be holes in your predictions. Ah! Says the economist, that's not all that important as long as we hit the main causal factors because we're looking for an aggregate prediction anyway....and he's right. Except that many of these aggregate predictions have no bearing on the individual outcomes...good or bad. Not only that, it might not even be a good predictor due to localized variation (there's a No-Pizza ordinance in XYZ county, therefore no pizza shops). Yet, the prediction, because it is true in aggregate, may be used to support further public policy.
So beware of using large population predictions to justify (or predict) individual experience. And be wary of the same thing being handed to you by the media/government/corperate complex.
End of rant.