In my last post, I noted the apparent contradiction between supply curves and learning curves. Both display a relationship between cost/price and quantity, but one displays a positive correlation between the two while the other shows a negative one. One variable that is missing from both graphs is time. Learning happens over an extended period, but what about supply? If we consider the short-run, the supply curve can possibly be viewed as a moment in time, that is time is fixed. Armen Alchian, a prominent RAND and UCLA economist in the 20th century, analyzed aircraft production data during World War II. He was aware of the apparent contradiction between learning and supply curves, and resolved it by stating that the supply curve should be viewed as a rate curve. The rate curve also has two variables – quantity and price/cost, but the rate curve is the rate at which production occurs. While learning occurs over the total volume of output, rate is the amount produced for a given period. The rate curve may be more in line with the traditional notion of supply in economics. Thus, the apparent contradiction between the learning and supply curves is more apparent than real.
Turning to the rate curve, however, research indicates that cost is often a decreasing function of the rate of production if the rate of production is below the system capacity. In aerospace and defense, for example, production rates are typically well below capacity. This means there is a significant fixed cost that when amortized across more units in a given time period, causes the average unit cost to decrease, rather than increase, as a function of the production rate. For example, a 2016 NASA study found an 84% learning curve and a 60% rate curve for Shuttle External Tank production. Even accounting for learning the data indicated that every time the production rate doubled, the unit cost dropped by 40%.
This phenomenon is not limited to defense and aerospace. A 1952 paper that surveyed over 300 businesses on how they viewed the relationship between unit cost and capacity found that approximately 95% chose downward sloping curves (Eiteman and Guthrie, “The Shape of the Average Cost Curve,” American Economic Review). In the late 1990s former vice-chairman of the Federal Reserve Alan Blinder published a book that included a survey of 200 mid-size and big businesses in which he reported that “only 11 percent of GDP is produced under conditions of rising marginal cost.” He further stated: “Firms report having very high fixed costs – roughly 40 percent of total costs on average. And many more companies state that they have falling, rather than rising, marginal cost curves. While there are reasons to wonder whether respondents interpreted these questions about costs correctly, their answers paint an image of the cost structure of the typical firm that is very different from the one immortalized in textbooks.” (Blinder, Asking About Prices, 1998)
Regardless of how quantity is interpreted on the traditional supply curve, whether as the total quantity of production or as a rate of production, the empirical evidence does not support the traditional textbook depiction of supply curves that always slope upward. The graph below compares how businesses view supply (based on the curve that 60% chose in the 1952 study) with how textbooks represent supply.
