[The entrepreneur] would (so it appears) plan a stream such as BB, which would rise while equipment was being adjusted to the new conditions, but would probably settle down in the end to a new 'equilibrium'.The analysis next turns to the various sources of the total effect of the price increase. Hicks argues that with elasticity of expectations equal to 1, the total effect "is compounded out of" two types of effects: the effect of a rise in the current price (with expected prices remaining unchanged), and the effects of a rise in expected price at each particular point in time (with the current price and other expected prices remaining unchanged). His exposition considers the second type first.
In supposing that an increase in the price of the entrepreneur's product is expected at the date M, the discussion examines two types of consequences of this expectation. The first type involves the entrepreneur substituting resources that had been intended to be used toward production at other times, either earlier or later, or possibly both, "in order to have as much as possible ready at the critical date." Various technical characteristics of the product and the equipment used to produce it will affect how easily substitution can be applied. Among such characteristics, the author lists the durability of the product, the durability of the inputs to production, the initial quantity of available inputs, and so on. He concludes that "the general shape of the output stream which will be planned" in this type of situation is that shown by the curve ACA'.
The second type of consequence of the expectation of a price increase will be more pronounced when there is less opportunity for substitution. This might be the case, for example, when "the product is not durable, and the materials which go to make it are not durable." To meet the expected price increase, the entrepreneur might invest in additional production equipment, which is itself durable. In this case, "the existence of the equipment will then facilitate increased output at other dates as well. This is the case of complementarity over time." In this case, the stream of planned outputs will have a form similar to that of AD.
For the case of a rise in the current price that is not expected to persist, there will typically not be enough time to install additional equipment, so the complementarity effect will be small. Substitution effects can still occur, but they can only involve substitution of resources that were intended to be used for future output. If we are truly talking about a rise in the current price (for which there is no advance notice), then no substitution could be done in the past in anticipation. If there is a substitution effect, the planned output stream will have the form shown by the curve EA'.
