Harold Hotelling, a Mathematician from Stanford University in March 1929, in his seminal paper, “Stability in Competition”, unraveled the secrets of “excessive similarities of competing commodities”, which effectively explains why vegetable sellers or fish sellers or other merchants of a commodity in a market are in one place instead being far apart, or for that matter what is the economics behind a super market against widely dispersed stalls all over the city. For that matter Hotelling provided the reason why it makes a great strategy for a Democratic Presidential candidate to be as close to center as for a Republican candidate, thus they mitigate their risks of going too far away from what the majority of the electorate would like (similar is better than being far away); too much right off center or too much left makes a losing case by this argument.
Hotelling was dealing with perfect competition and with commodities with replicable properties. He was more a Statistician; we should give credit to him for his proof for perfect competitiveness that minimizes social costs.
If we take the metaphor of transportation cost as the key differentiator between two commodities then consumers would try to minimize the total cost of transportation and thus it would make great sense for the shops to be as close to each other than being as far apart to remain competitive. To take the point as Hotelling constructed it, suppose there are consumers situated on a straight line on a road and there is a duopoly of two players trying to sell the same product on that line. It would make great sense for the two players to be either at the two extremes of the line or at the center or at any equidistant position from the center of the line. This way the two players would ensure that the transportation cost delta between them would be zero, thus the differentiator gets negated.
Imagine an ice-cream vendor trying to sell ice-cream on a beach and takes a position at the center of the beach. The next ice-cream vendor would have no other option but to stay as close to the first as possible so that he is able to negate the differentiator (which is the transportation cost of the consumer to approach him, which is actually the decrement of utility resulting from the actual commodity being in a different place).
The purpose of maximizing competitive advantage in this case is to lower the differentiation costs to zero. The proof of this was provided by Hotelling and as Calculus of variation allowed this to be done, the matter ended there. Hotelling never used the term ‘differentiation’ in his paper, by the way and it was only later that this got somewhat framed by Edward Chamberlain in 1933, just four years later, in his treatise, ‘Theory of Monopolistic Competition’.
What Hotelling missed out to see was that it was possible for monopolies to be created where certain product attributes could be made such that were not possible to be imitated by the others. New products, applications that could be later patented allowed this to happen. Chamberlain did not believe that it was a failure of the market as it was crafted on imperfect competition, which many argued was a departure from the pareto optimality. In fact he believed that it increased the economic welfare by creating as much product diversity as the society was willing to pay for.
Indeed today one would agree with him that the likes of Google, Facebook or other such monopolies actually create economic welfare in spite of being monopolistic in nature (monopoly could be a bad word, the better way to represent this would be to say that they own dominant position in their product and service segment). However Chamberlain’s work got a stiff opposition from the Chicago School and particularly got over-shadowed by the work of Joan Robinson although the central tenet of his argument remains valid even today that in the short run equilibrium the monopolistically competitive firm results in economic profit but with free entry the long run equilibrium moves towards normal profit and excess capacity. The two differed in their explanation and while Robinson moved to term this short term phenomenon as ‘exploitation’, Chamberlain explained that it was due to product strength, promotion and advertising, which he termed as ‘diversity’.
It is an irony that the Nobel Prize on this subject went to Vernon Smith in 2002, who actually got all the inspiration from Chamberlain’s seminal paper, ‘An experimentally imperfect market’.
Why would there be such a gap of fifty years needed to understand Chamberlain’s imperfect market? It is because the economists could not understand differentiation for many long years and it did not fit the mathematical models they had constructed for oligopolistic competition; it was also because the economists could not accept the excess capacity conclusion as it clashed with the general equilibrium theory. Although Lancaster in 1975 came up with the understanding that there was a socially optimal degree of product differentiation, but it never could make a concrete case for establishing this degree.
It had to be the branch of Management Sciences, developed by Michael Porter that this got studied and deliberated that allowed firms to focus on as the most potent discipline for making economic profits, which is product diversity.
The perfect blend of monopoly and competition was at the core of Chamberlain’s belief, which would eventually achieve welfare objectives of the society. The equilibrium condition that prices should reflect marginal costs was a more prescriptive (desirable social objective) denouement, which Chamberlain believed was never a limiting condition that market participants could be subjected to. He said, “Whenever there is a need for a diversity of any product, pure competition turns out to be not the ideal, but a departure from it. Marginal cost pricing no longer holds as a principle of welfare economics, nor the minimum point on the cost curve to be associated with the ideal.”
Today Google, Facebook, Tesla, Uber, Amazon, and all those disrupters who have created diversity, or new needs or new platforms, traditional welfare economics fails to correlate with their models to explain pricing, costs and utility.
Thanks to Jean Tirole, we came to some conclusion on two-sided markets. But differentiation is yet to be fully explained by economists in furthering the cause of social welfare. At least in this area they have some work in progress.