As a reference point, we analyze the case where the consumers
are fully informed about their individual values of
, the
producer knows the distribution
, and the producer
and the consumers act so as to maximize their expected gain.
Integrating g, we obtain the cumulative distribution
. From Eq. 6
we can compute the conditional distribution
The average valuation for this conditional
distribution is
.
Using Eqs. 5 and 9, we obtain
the expected number of purchased articles and the expected surplus (assuming the
consumer subscribes):
To compute the producer's expected profit as a function of
and F, we can substitute Eqs. 12
into Eq.10,
which yields:
where
is defined as
and
is defined as the unique solution to
.
Eq. 13 can be visualized as a profit landscape
in which the expected profit is plotted as a function of F and
. It is convenient to define a normalized expected
profit
and a normalized fee f = F/N.
Fig. 1
illustrates the landscape for two
different production costs:
and
.
Figure 1: Profit landscape
.
h is a uniform distribution with
,
. a) Production cost
.
b) Production cost
.
In each such landscape, there are two ridges. The lower ridge,
which demarcates the boundary beyond which the producer
attracts no consumers and thus makes no profit, is defined
by
.
The upper ridge is described by a piecewise joining
of two nonlinear curves, the simpler one being described
by the relation
.
This portion of the ridge has the following characteristics:
to be
approximately 0.28,
so that Fig. 1a (
.
Figure 2 shows the dependence of the
optimal f and
upon the production cost
.
There is a discontinuous derivative at
,
due to the switchover between the two nonlinear curves
that define the upper ridge in the landscape. As one
might guess, the profit
decreases monotonically
with the production cost.
The proportion m of consumers that subscribe is 1
for all
; for
exceeding this threshold the proportion of subscribers
is strictly less than one, and is given by
.
Figure 2: Optimal price-per-item
, normalized subscription fee f,
normalized profit
, and fraction of subscribers m vs.
item production cost
,
where h is a uniform distribution with
,
.