Wednesday, September 4, 2013

The artificial separation of expectations

Imagine you are buying a car. What might you consider?

  • Price of the car (financing, quantity available)
  • Dimensions (number of seats, cargo capacity, urban vs rural use)
  • Amenities (leather seats, A/C)
  • Gas mileage (weighted by cost of gas)
  • Maintenance costs
  • Resale value
  • Traffic congestion
  • Policy changes (registration fees, congestion pricing, tolls)
Why separate these pieces of information into sources of "current value" and sources of "expected value"? It is true this seems more relevant for an options contract:
  • Market value
  • Strike price
However, it is important to realize in this case the strike price anchors the future expected value (measured as a difference from the strike price). The two measures are completely intertwined -- the market value as a piece of information has no meaning without the strike price. The discounted expected value is basically the measure that takes into account both these pieces of information. The success of the Black-Scholes model lies in its ability to account for the supply and demand price of the stock simultaneously with the expected value of the derivative.

There are behavioral factors (like money illusion) that show there are differences between "expected value" and "current value" and people seem to be given to hyperbolic discounting. But this is an even more serious problem for rational expectations where the deviation from the model price is assumed to be an unbiased random fluctuation. This seems all the more reason to just lump expected value into the other reasons people buy (or don't) at the current price.

The distinction between sources of expected value (based on market psychology and economic models) and sources of current value (based on the invisible hand and the price on an open market) is artificial. They both represent pieces of information used to determine if an economic agent should buy at the current price or not. We already assume people randomly buy at the "incorrect" price for reasons we don't model (I will buy a pack of post-its from my local grocery store even though they are cheaper at an office supply store or online due to spatial, temporal and accounting convenience). That price I paid enters into the "going rate" of post-its. It communicates information to the supply as well as any other purchase.

1 comment:

  1. A similar argument is discussed here

    Especially point 9:

    "There is no clear distinction between expectations and fundamentals, because expectations are fundamentals."


Comments are welcome. Please see the Moderation and comment policy.

Also, try to avoid the use of dollar signs as they interfere with my setup of mathjax. I left it set up that way because I think this is funny for an economics blog. You can use € or £ instead.