23
individuals unique to each other, not easily replaced by others with similar characteristics – meaning
that market power is everywhere as well.
The first fundamental theorem of welfare economics is often presented as saying that the market
equilibrium is Pareto efficient, provided that certain assumptions are fulfilled. Given Proposition 1
above, an alternative and perhaps more helpful statement is the following: the first fundamental
theorem of welfare economics establishes that to guarantee market equilibrium efficiency,
extremely strict assumptions are needed.
Social interaction in markets
Economists’ notion of perfect competition was hardly contrived of to analyze the private family
sphere, with which my fiction story is largely preoccupied. Note, however, that this is not the source
of the story’s absurdity. Had the tale taken place in a market context, such as the factory floor or at
the hairdresser’s, it would presumably have become equally bizarre.
Absurdities arise because external effects, asymmetric information and other market failures are
integral parts of normal human interaction, shaping behavior wherever people meet and relate to
each other – within and outside of markets. The problems discussed above are thus likely to be
present in most markets, but may be of particular importance where human interaction is an
essential part of the transaction – e.g., in labor markets and markets for personal services.
Social interaction is involved in all socially learnt and culture-specific behaviors, of which many affect
markets: leadership styles, teaching styles, clothing styles, dietary customs, and so forth. Examples of
economic implications of social interaction are numerous, including conspicuous consumption
(Veblen 1899); housing market segregation (Schelling 1978); fair wage-setting (Akerlof 1980, Bewley
1998); crowding phenomena like waiting lines in restaurants (Becker 1991); customs, fads and
subcultures (Bernheim 1994); cooperation and revenge (Rabin 1993, Sobel 2005); employee
motivation (Benabou and Tirole 2003, Dur and Delfgaauw 2008, Brekke and Nyborg 2010); social
norms, conventions and coordination equilibria (Bicchieri 2006, Young 2015, Nyborg et al. 2016), to
mention some.
One aspect of social interaction that still seems understudied in economics is the phenomenon of
bonding. Precisely because it is vital to human well-being, social bonding is likely to lead to severe
inflexibility in markets such as labor and housing markets. For example, befriending one’s colleagues
bestows market power on one’s employer; similarly, landlords gain market power if their tenants, or
tenants’ children, befriend the tenants next door.
Final remarks
The fiction story about the perfectly competitive market comprising Part II of this essay proceeds
much like a standard economic theory paper: it specifies its assumptions early on, then exploring the
logical implications of those assumptions. This yields several rather surprising insights, some of which
may have been difficult to see using a more conventional approach. In particular, if trade can occur
at any time, and deliberate, potentially welfare-relevant learning is feasible, no perfectly competitive
market can exist.
Some may object that taking the assumptions of no market failure literally, like I have done here, is
unreasonable. For example, some of the external effects caused by normal social interaction may be
minor, only marginally relevant to markets. Nevertheless, adopting a flexible view of one’s
assumptions is not the usual standard in economic theory. Modifications of assumptions should be
explicit; complications cannot be dismissed by hoping that their importance is limited.