Economics vs Finance
It seems to me that one of the main sources of the current crisis is a fundamental flaw in finance theory, relative to economic theory.
In most economic models, prices are derived in terms of underlying fundamentals - cost of production, consumer utility, etc.,. This includes the most basic of all economic models, the intersection of supply and demand curves.
In contrast, many (most?) successful and widely used models in finance express the price (value) of financial assets in terms of other financial assets. The CAPM and APT show how to fairly price assets relative to the value of the total market or to aggregate factors. The Black-Scholes formula shows how to price options relative to the price of the underlying assets. Tools like these were applied and generalized to construct and price ever more exotic derivatives, like the now toxic collateralized debt obligations.
The problem with these financial models is that they don't identify or rule out massive, systemic bubbles. If a stock is not mispriced according the CAPM (has a zero alpha), and the price of (or more precisely the returns associated with) every stock including that one increases 1,000%, or 10,000% then the CAPM will still say the stock is correctly priced, because the CAPM only judges the return on an asset relative to the returns on a basket of other assets.
In exactly this way, the financial models used to construct the infamous mortgage backed securities appeared to work perfectly (until the real world intervened), because no matter how overpriced actual houses became in any real economic sense, the models said the securities were appropriately priced relative to the (insane) housing prices.
This suggests to me that what the next generation of financial models require is some economics, that is, a deep way to incorporate economic fundamentals, to avoid (or in the case of wall street, to profit by exploiting) not just relative mispricing, but absolute mispricing of financial assets.
In most economic models, prices are derived in terms of underlying fundamentals - cost of production, consumer utility, etc.,. This includes the most basic of all economic models, the intersection of supply and demand curves.
In contrast, many (most?) successful and widely used models in finance express the price (value) of financial assets in terms of other financial assets. The CAPM and APT show how to fairly price assets relative to the value of the total market or to aggregate factors. The Black-Scholes formula shows how to price options relative to the price of the underlying assets. Tools like these were applied and generalized to construct and price ever more exotic derivatives, like the now toxic collateralized debt obligations.
The problem with these financial models is that they don't identify or rule out massive, systemic bubbles. If a stock is not mispriced according the CAPM (has a zero alpha), and the price of (or more precisely the returns associated with) every stock including that one increases 1,000%, or 10,000% then the CAPM will still say the stock is correctly priced, because the CAPM only judges the return on an asset relative to the returns on a basket of other assets.
In exactly this way, the financial models used to construct the infamous mortgage backed securities appeared to work perfectly (until the real world intervened), because no matter how overpriced actual houses became in any real economic sense, the models said the securities were appropriately priced relative to the (insane) housing prices.
This suggests to me that what the next generation of financial models require is some economics, that is, a deep way to incorporate economic fundamentals, to avoid (or in the case of wall street, to profit by exploiting) not just relative mispricing, but absolute mispricing of financial assets.