Congestion and Network Externalities in the Short Run Pricing of Information System Services

Transfer pricing provides management with an efficient tool for coordinating the use of information system services. In practice these services, if they are charged out to end users, are priced via an allocation of historical costs based on individual usage, or are assigned a price based on the fair market value of outside alternatives. Both approaches encourage suboptimal usage of information systems resources: they ignore three constituents of optimal transfer prices: (1) the reaction of other users to any change in a given user's consumption of service; (2) the technical response to additional use of an information system service, in the form of congestion and network externalities; and (3) end-user preferences for the information system service. Together, these influences on demand contribute to an externality. Two types of externalities are significant in information systems---congestion externalities decrease, and network externalities increase the utility of services. This research provides a complete model for computing optimal transfer price which corrects several existing problems in information system service valuation. When full absorption cost allocations contain significant fixed, precommitted or sunk cost components, it is not possible to balance the budget. Direct costing improves on full absorption costing, but misspecifies costs when demand is nonlinear. To balance the budget with a minimum of disutility to end users management must resort to second-best pricing. Inefficiency arises from two sources when externalities are ignored in transfer pricing. End users cannot plan service usage, and thus encounter unanticipated congestion which lowers morale and efficiency, and increases costly expediting. Where congestion effects are highly nonlinear, congestion may effectively shut down the system. On the other hand, incorrect transfer prices can motivate usage too low to take advantage of network benefits. End users may confuse externality effects and intrinsic value of an information system service. In these cases, the price and performance of outside alternatives exert a complex and subtle influence on use and pricing of services. Where there exist identical outside and inside processing alternatives, and transfer pricing does not reflect this, there is a significant tendency to overprice and underutilize internal services. In this case there may be a tendency to outsource when processing should be kept in-house. Future markets for information systems will favor technologies which exhibit strong network externalities, and pricing of these technologies will become increasingly complex because of the interrelationship of hardware, software and data originated externalities. The models provided in this research will allow management to take advantage of these opportunities.