William Niskanen argued that the characteristics of the legislative review process imply that bureaus are allowed to produce an output which is less efficient than the legislator would allow [4; 5]. The main points Niskanen raised in this regard centered on the facts that: (i) A committee, not the whole conducts the review process for each bureau; and (ii) More important, at least in the United States, most committees are dominated by legislators who have higher demands for the services reviewed than the median in the whole and the committee decisions are very seldom amended or reversed by the whole legislature [5, 624-25]. Niskanen observed that the review committee prefers a greater output than that preferred by a legislator [5, 625-26]. Since bureaucratic output is inefficiently large in the Niskanen model of bureaucratic production, the implication is that demand committee review allows relatively more inefficiency than the representative legislator would condone. Niskanen assumed that all legislators wish to maximize their political support (votes). Such legislators would not allow an increase in inefficiency without some offsetting benefit, since some voters have a negative reaction to this inefficiency (non-consuming taxpayers). Yet, in the Niskanen model the bureau captures all the consumer surplus associated with the increase in output allowed by high committee review. Consumers of the bureau's output may appear to benefit since they consume more. However, they gain no additional surplus so there is no increase in consumer (voter) welfare. There appears to be nothing positive to offset the loss in votes resulting from inefficient production. Under majority rule decision making, it is often argued that the desires of the median voter will be met. The legislator is the median voter in a legislative assembly. Therefore, it seems that this median legislator's perception of marginal political returns and marginal political costs should dictate the allocation of resources in the public sector.