In my recent blog post Neoclassical Economics: The Cult of Marginality, I critiqued the use of marginalism in economics. My post received some good feedback, and I am writing this post in response. One response was that I was confusing the short run with the long run, so I decided to do some more thinking on the subject. In my previous post on this subject, I was attacking the idea of marginalism in neoclassical economics. This idea is pervasive throughout much of the prevailing school of economics, which is known as the neoclassical school. It is so foundational that this school is called the marginalist school of economics. The devotion of the school to this concept is so extreme and so misguided that I call it a cult – I call it a ‘cult of marginality’ as a play on the expression ‘cult of personality.’ One of pillars of economics that rests upon the foundation of marginalism is the notion of the supply curve , which is the price at which firms are willing to provide a given quantity of a specific good. The supply curve shows not only a positive relationship between quantity and price, but a curve that is accelerating upward. As quantity increases, price increases, and the rate at which price is increasing is also increasing. For example, if three quantities are 1,2,and 3 and their associated prices are $5,$10, and $20, then the increase in price from the first to second unit is $5, but the increase in price from the second to the third unit is $10, which is double the increase from the first to the second unit. Neoclassical economics also assumes that there is perfect competition in the economy. These lean and mean companies are willing to accept no surplus profit (beyond an opportunity cost), so price equals cost. This reflects the idea in economics that it costs more and more to produce a good as the quantity increases. This is known as the ‘Law’ of Diminishing Returns . See the graph below for a notional illustration of a supply curve.
Credit for the supply curve goes back to the British economist Alfred Marshall, one of the founders of the neoclassical school. He published a landmark textbook in 1890 titled Principles of Economics that expounded the principles of supply and demand, and related supply to the cost of production. A more mathematical treatment was developed by the University of Chicago economist Jacob Viner, who in 1931 published a paper titled “Costs Curves and Supply Curves” that discussed the short-run and the long-run. This is about the extent of the treatment of the dimension of time in economics – it has changed little since. Viner defined the short-run as “a period which is long enough to permit of any desired change of output technologically possible without altering the scale of plant, but which is not long enough to permit of any adjustment of scale of plant.” In the short run, you cannot make the plant bigger, but you can add labor or machinery. In the long run, you can take advantage of technology advances, you can build bigger plants, etc. Pretty much anything is possible.
A few years later, the engineer Theodore P. Wright discovered, in his work with production in the aircraft industry, that costs actually decrease as the quantity produced increases. He called this phenomenon learning. The general rule of thumb is that costs decline by a constant percentage every time production doubles. This has been found to be a universal phenomenon in manufacturing – missiles, semiconductors, satellites, refrigerators, etc. This is illustrated in the figure below.
Note that there is nothing in the concept of learning that violates Viner’s definition of the short run. The apparent contradiction was ignored by economists for several decades, until one of them ran head first into reality. Armen Alchian was a prominent economist at UCLA who also did some work for the RAND Corporation, a government think tank, in the 1940s to 1960s. There he discovered the learning phenomenon when looking at production cost data for aircraft. In an attempt to reconcile this contradiction, Alchian came up with the idea of supply as rate rather than quantity. He developed a multivariate cost function that included both total quantity (Q) and the rate of production (q). He posited that the part of cost reductions, the learning phenomenon, was due to increases in Q. The supply curve related to changes in q. Alchian posited that diminishing marginal returns was still present with respect to changes in cost relative to changes in q. To try to salvage the ‘Law’ of Diminishing Returns Alchian had to change the notion of the short run and make it an even shorter run.
This seems like a nice notion. I would like for it to be true. However, I have also modeled rate effects for production costs, and they do not exhibit diminishing returns either. It is common in the estimation of production costs to account for both learning (Alchian’s Q) and rate (Alchian’s q). Rate, in actual practice, exhibits the same shape as the learning curve. This is because, largely due to uncertain demand, producers have excess capacity. There is a large fixed cost. You can produce a smaller or larger quantity for very little change in cost. The marginal cost to produce an additional unit is typically modeled as a fixed amount. This amounts to a flat marginal cost curve as a function of rate. See the graph below for an illustration.
As you can see, there are no diminishing returns here either. In theory, if the rate were increased enough, you would see increases in marginal cost. But that’s the theory, not the reality. In reality, with large fixed costs as a result of excess capacity, most manufacturers are not producing anywhere close to a rate that would result in an increase in variable costs. And with a large fixed cost, pricing will not occur on the marginal cost line (note that in the graph above, I removed the reference to price on the vertical axis). Alchian’s attempt to salvage upward sloping supply curves did not work. At best, the supply curve for a product can be conceived as a horizontal line. The neoclassical theory of the firm does not model reality regardless of how its defenders change the notion of the short run.