ISF WP 2013-3 - page 9

9(34)
Producer Behavior
Any analysis of the costs and benefits of public procurement rules
presupposes behavioral assumptions concerning both producers and
procurement officers. The next three subsections justify the use of a non-
strategic model of producer behavior, different from the standard model
used in the academic literature.
Many factors affect the willingness of producers to take part in procurement
competitions as well as their behavior once they decide to participate.
Some of these factors are related to the general background or
environment, while others to the specific project. Among background
factors, the
field of competence
of the producer is of course basic.
Size
may be important, and sometimes minimum size (turnover, number of
employees, etc.) is prescribed in procurement background documents. The
company may have a general
policy
that affects its choices in procurement
matters, for instance concerning what types of projects to engage in,
pricing, and cooperation. The
order intake
will obviously be important both
for the interest to participate and for pricing. Given that any tender is
presented under some degree of uncertainty, the
attitude to risk
may also
be of importance. In general terms, large companies can afford to be more
risk-prone than can their smaller competitors.
Among project-specific factors, the
assessed profit
from the project is
obviously central to the company’s decision to participate. Profit is
determined by the
costs
and the
price
obtained. Costs are always
uncertain, and these are normally estimated based on key numbers
(cost per kilometer of road, per consultant-week, etc.). Specific conditions
unknown to the tenderer may give rise to unpleasant surprises. In cases
when the procuring entity bases its decision on both
price and
quality
, the
producer must also decide how different alternatives will fare depending
on the quality-to-price evaluation formula. The identity of the
customer
may be important; some customers may be useful for reference in future
marketing efforts. Finally,
potential
competitors
affect the tender. This is
also an uncertain factor given that neither the number nor the identity of
other tenderers is known when the producer makes its choices.
The risk of
cartels
is always present. During the work on the present
study, a cartel was suspected by the author and reported to the Swedish
Competition Authority. This eventually resulted in a fine for the four
companies involved.
The list is not exhaustive, but nonetheless offers a reasonably adequate
picture of the diversity of factors involved.
Standard Model
The standard model used in the analysis of procurement is based on
auction theory, which is a branch of game theory (see, for instance,
Klemperer (2004), Menezes and Monteiro (2007), and Milgrom (2004)).
The auction model assumes a
known number
(
n
) of participants, denoted
x
i
, i = 1, 2, ..., n. Participants are assumed to be
risk-neutral
. The valuation
v
of what is offered varies by participant and is denoted
v
i
. The
distribution
of these valuations is also assumed to be known and is denoted by
F(v)
;
the corresponding frequency function is
f(v)
. A solution, if it exists, is
1,2,3,4,5,6,7,8 10,11,12,13,14,15,16,17,18,19,...34
Powered by FlippingBook