Disclaimer. Don't rely on these old notes in lieu of reading the literature, but they can jog your memory. As a grad student long ago, my peers and I collaborated to write and exchange summaries of political science research. I posted them to a wiki-style website. "Wikisum" is now dead but archived here. I cannot vouch for these notes' accuracy, nor can I say who wrote them.
Niskanen. 1975. Bureaucrats and politicians. Journal of Law and Economics 18 (December): 617-43.
Although Congress has Constitutional responsibility for many policy areas, it has abdicated this responsibility to largely independent bureaucracies.
Niskanen argues that the size of the discretionary budget (Y) reflects the reward structure in the bureaucratic environment (X)--not Congressional priorities.
Along with Lowi, Niskanen became the target of later works seeking to demonstrate that Congress does, in fact, control the bureaucracy. See especially McCubbins and Schwarts (1984) and Kiewiet and McCubbins (1991).
Niskanen begins by showing that bureaucrats maximize personal utility (wages and perks) by maximizing the agency's budget. Actually, he corrects this statement. Instead: bureaucrats maximize objectives defined in terms of the agency's discretionary budget. (The discretionary budget is the difference between the budget Congress gives you and the cost to the agency of producing its output).
Niskanen then argues that legislators have ways to monitor bureaucracies. However, they are costly. Several reasons:
The result: legislatures will use these mechanisms of control only to the point where their marginal value equals their marginal cost.
For Niskanen, the primary actors are bureaucrats and members of relevant Congressional committees. Bureaucrats seek to maximize income and perks. Committee members maximize reelection odds; since the politicians on the committee are there by choice, they will tend to be interested in an inefficiently large output from agencies under their jurisdiction (so they are willing to give more money than the full Congress would, but they expect it to be used as efficiently as the full Congress would).
Result of this interaction: bureaucrats want an oversized budget. Their overseers in Congressional committees want an oversized budget (relative to what the Congressional median would want). So you get an equilibrium.
Implication for efficiency: Congress will overconsume its common pool resource (tax dollars), leading to a growing spending deficit.
As a result, we see weak attempts to prevent this behavior, such as administrative law (solves the efficiency problem, not the common pool problem), line-item vetoes (helps the common pool problem), etc.
Research on similar subjects