Dynamic Explanation of Industry Structure and Performance
Cotterill
vanishes into smoke. (Franklin as quoted in V.S. Clark,
1949, p. 152)
The division of labor is not a quaint practice of
eighteenth-century pin factories; it is a fundamental
principle of economic organization. In the present
instance, the organization of food industries depends
upon our human capital infrastructure as well as the
physical infrastructure of the economy. These change
over time.
George Stigler (1951) is the only economist to
squarely address the issue of vertical market channel
organization within the confines of neoclassical theory
of the firm and market equilibrium. Perry (1989)
acknowledges that Stigler's effort was limited to his
short 1951 article and 3 or 4 subsequent pieces have
attempted to test or expand Stigler's theory.
Stigler introduces his theory by explaining that the
firm is portioned among production processes rather
than the usual input market configuration i.e. the
textbook example of production isoquants in capital and
labor space. In his basic model he ignores the possibility
that costs for one process are related to costs of the other
distinct processes. This means that one can derive a cost
function for each process that is only a function of
output. Finally he assumes fixed proportion (but not
constant returns to scale) production so that one can
draw all cost functions on one graph with final output as
the quantity index on the X axis.
Figure 1, reproduced from Stigler (1951) illustrates
the theory for a firm with 3 distinct production
processes. Note that the U-shaped average cost for the
firm is the sum of the 3 process cost functions. Process
Y1 has increasing returns. Process Y2 has decreasing
returns, and process Y3 has both to produce U-shaped
curve. A critical question is when will Y1 splinter from
this firm, and presumably other firms in this industry, to
become a separate industry?
Stigler answers that at a given time this process may
be too small to support a specialized firm or firms. He
states:
The sales of the product may be too small to support a
specialized merchant; the output of a by-product may be
too small to support a specialized fabricator; the demand
for market information may be too small to support a
trade journal. The firm must the perform these functions
for itself. (Stigler, 1951, p. 188)
This proves Adam Smith's theorem that the division of
labor is limited by the size of the market.
As the economy, and in particular this industry
grows, the magnitude of the function subject to
increasing returns may become large enough to permit a
firm to specialize in producing it. This new firm may
initially be a monopoly but its limit price would be
determined by the old industry's ability to revert to in-
house production. With growth over time output
expands until process Y1 also experiences decreasing
returns and then one might see entry into this new
industry and a trend towards a competitive structure.
This is the reasoning for Stigler's implicit competitive
economy.6 Imperfect competition exists, subject to a
limit price, only until economic growth deconcentrates
the new industry. Lest one think we not need an
expanded theory, when have we observed such industrial
deconcentration with growth?
Note in Figure 1 that if the spin off produces Y1 at a
cost equal to the horizontal dotted line then the firm that
does processes Y3 and Y2 enjoys the new lower average
cost curve given by the dotted shift in AC. The spin off
not only reduces costs, it reduces optimal scale. Again a
curious prediction, counter to what we have experienced
in this century. Economic growth has not led to lower
optimum scale in industry as spin-offs have created new
and large, relative to the market, optimum scale
industries. Consider for example the optimum sized
farm over the 20th century.
Stigler also notes that outsourcing of production
could also occur for Y2, the process with decreasing
returns. In this case economic growth and spin off
increases optimal scale. An example of this effect is
food manufacturing companies that have dismantled
brand marketing units and outsourced individual
components such as focus group research, and
econometric analysis of demand to smaller boutique
firms. Outsourcing of advertising programs may go this
way but more often it seems to be subject to increasing
returns because it is usually awarded to large-scale
advertising agencies rather than boutiques.
But for the fleeting possibility of monopoly, Stigler's
theory focuses only on costs and implicitly assumes that
the least cost combination determines the market channel
structure. Baligh and Richartz (1967) and other logistics
oriented analysts have expanded this functional cost
based analysis, but none generalize the model to
6 Figure 1 is an exact reproduction of Stigler's figure. Note
that price, not dollars or costs, is on the Y axis. Stigler
clearly maintains that these cost curves determine price, but he
does not explicitly include a demand curve, nor does this say
how these cost curves determine industry supply. No
equilibrium price is identified in the figure. Stigler's theory is
incomplete unless one provides his implicit competitive
markets assumption and then identifies price as the LR
equilibrium price that occurs at the minimum point of the
average cost curve.
Food Marketing Policy Center Research Report No. 53