Improving Oversight: Congress's Endless Task
Joel D. Aberbach, University of
California - Los Angeles
For many years oversight was said to be Congress's neglected function because members lacked adequate incentives to perform what was said to be a demanding and politically unrewarding task. They would, analysts and candid insiders said, rather pass legislation establishing new programs or perform other tasks that would bring them kudos from constituents or interest groups than spend time reviewing administrative activities or evaluating the effectiveness and efficiency of programs.
With the growth of the federal government in the 1930s and 1940s, the need for oversight became more and more obvious to many members and observers of Congress. Reforms were enacted, such as the Legislative Reorganization Act of 1946 and amendments to it passed in 1970, which explicitly recognized the importance of the oversight function and encouraged an increase in oversight activity.
The seeming failure of these acts vexed many in Congress, and stimulated members and academics to ask why oversight remained a relatively infrequent endeavor. Members kept insisting that Congress should do more. Indeed, at a hearing in 1973, a fascinating exchange occurred between Professor Richard Fenno, a leading scholar of Congress, and Representative John Culver of Iowa. In response to questions about why past reforms had apparently had so little effect, Fenno opined: "When the incentive isn't there, you are simply not going to get oversight." Representative Culver was most unhappy with this sort of realpolitik explanation. Sounding more like an academic advocate of reform than a hard-nosed congressional veteran, he said: "I am really at a loss to follow you when you say there is not adequate political incentive. I think that there should be. After all, [congressional oversight] is our primary responsibility. If there is any disenchantment in the country, it derives from the validity of the thought that in this respect, Congress is pathetic."1
A study I later did on the subject, Keeping a Watchful Eye: The Politics of Congressional Oversight (Brookings, 1990), showed that starting just a few years before the 1973 hearings and accelerating afterwards Congress actually began to do more oversight. In fact, the study showed that when the situation was right, when factors existed such as widespread public disenchantment with government, relative resource scarcity, and a high level of tension between the legislative and executive branches, Congress could and did do plenty of oversight. Oversight comes to the fore at such times precisely because of the point made by Professor Fenno and others; members then have a political incentive to do it.
The data for Keeping a Watchful Eye tracked oversight activity into the early 1980s. Recent evidence demonstrates that Congress continues to oversee actively, although it also indicates the troubling fact that Congress is neglecting one important venue for oversight that it previously used to great effect. For analysis and interpretation of these data, I immodestly urge you to read my forthcoming article in Congress and the Presidency (Spring 2002). The problems reported in the study suggest significant avenues for continued inquiry on the quantity of oversight, but it is still fair to say that oversight, especially what I call primary-purpose oversight (committee hearings and meetings primarily devoted to oversight), is no longer Congress's neglected function and has not been neglected for many years. But this has had little impact on the level or intensity of complaints about congressional oversight. Why?
Oversight ProblemsMuch of what congressional and other critics complain about is the quality of oversight. It's not objective enough. It's not systematic enough. It lacks continuity. The complaints are consistent despite the continued high levels of oversight. For example, in his remarks to a congressional oversight workshop held in 1999, former Representative Lee H. Hamilton, one of the most respected members of the House during his tenure in that chamber, stressed that "Congress has given too much focus to personal investigations and possible scandals rather than programmatic review and a comprehensive assessment of which federal programs work and which don't." He listed a set of observations about what makes for "successful oversight." Oversight should be bipartisan, constructive in the sense that it puts aside "petty political motives," regular and systematic, comprehensive, coordinated, reflect the expertise of a professional body, be careful not to intervene excessively or fall prey to micro-management, involve members as well as staff, result in clear reports and follow-through by committees, and reflect the priority of the congressional leadership that this is the way to do things.2
Representative Hamilton was so admired in the House because he actually practiced what he preached to the workshop. Indeed, one of the major reasons he was held in such high esteem is that his behavior was unusual. More commonly, the average member is influenced to a greater extent by partisan concerns and the need to shore up his or her political base, and members have "agendas" that influence what they do and how they do it.
The fact is that Congress does not do most of its oversight the way standard evaluation texts might suggest. It is not objective. While Keeping a Watchful Eye documents that congressional committees use (or consult many analyses that use) techniques developed by program evaluators, they use these analyses in a political context. That's hardly surprising. Congress is a political body. Its members are the quintessential non-objective types. They are not there to prove scientifically what's right or wrong or to approach problems in a neutral manner, but to represent interests and to broker deals and, increasingly, to express values. They are fondest of things that reinforce the views they advocate, and tend to regard the other side's views as wrong-headed, wrong-hearted, or wrongly-conceived--often all three. They frequently make deals, of course, an important part of political life, but more in the interest of getting things done than because they have been converted.
What this means for oversight is that it is a part of the political process. In Keeping a Watchful Eye I characterized much of Congress's oversight behavior in the period I studied as taking place in an "advocacy context." What I meant by this was that oversight was conducted in an environment of support for the basic goals of programs and agencies. Congressional committees were often critical of government activities and of program and agency administrators, but they generally expressed their criticisms in a context of sympathy toward the agency's or program's basic charter or purpose. Congressional reviewers were predisposed to preserve the core of most government programs even when they felt it necessary to modify some aspect of policy or of agency behavior. The data demonstrated that Congress used many sources of information, including at times information generated by studies using the "scientific" techniques of program analysis, but this information was consumed in an advocacy context. The conclusion drawn was as follows:
Congress…is a political body filled with people who have preferences and who are given license to pursue those preferences through their committee assignments. The political nature of the body and its members' preferences do not inhibit, or even much limit, the occurrence of oversight in its many manifestations, but they tend to shape what occurs. Members of Congress want to know and do know more than one might naively suspect, but naturally enough they want to know more about some things than about others. And the information they get is viewed through their politically ground lenses and used for their own purposes.3
Keeping a Watchful Eye was written while the Democrats, the designers and supporters of most domestic programs, still held the upper hand in Congress. Republicans had begun their slash and burn attacks on federal programs, but the intensity level was still relatively low, especially within the committees where many Republican members became qualified (and sometimes enthusiastic -- the Republicans had some favorite programs of their own, especially in the national security and law enforcement areas) supporters of many aspects of the programs their committees authorized or appropriated. Advocacy oversight, to reiterate an important point, is not uncritical--indeed it is often quite critical--but the underlying assumption is that the program will (and should) continue, with mainly incremental changes to improve it. In this context, studies that start by accepting the goals of programs often make significant and even well received recommendations for restructuring or reform. But the underlying aim is to accomplish the program's goal in a better way. Such reviews are serious and sometimes quite critical, often promoting heated debate, but they are rarely life threatening. The question is not "Should government do this at all?" The question is "How can government do this better?" Or, in times of divided government, "Why are those nasty political appointees sabotaging programs?"
The 1994 elections brought Republican majorities in both houses of Congress. This heady change unsettled many conventions, including the advocacy context that had settled over Congress. Now the questions broadened out. In fact, the ultimate question--"Why is government doing this anyway?"--now found its place at the forefront of much committee oversight activity. And the criticisms of everyday activities often had a harsher ring than in past years.
A look, for example, at the activities report issued in 1999 by the House Committee and Education and the Workforce (the tellingly renamed post-1994 version of the old House Education and Labor Committee) is quite interesting. The report stresses that "current law is inadequate" in the area of the use of dues by unions for "political and other non-collective bargaining purposes."4 It praises work on a project begun in the 104th Congress (1995) to determine "what works and what's wasted in Federal education programs," with an overall finding that "the current federal role in education does not generally support…successful schools. Instead, it is uncoordinated and duplicative, and has produced little evidence of effectiveness."5 And on the question of the National Endowment for the Arts, the report approvingly endorses a conclusion of the Subcommittee on Oversight and Investigations: "The NEA does not have an appropriate federal role; it does not operate effectively and efficiently; and it does not follow congressional intent."6. At the same time, the report is effusive in praising the welfare reform effort that the committee helped shape in the 104th Congress, stating that it "continues to be a success."7
My point is not to say that oversight has declined in quality under the Republican majority or that every committee has sought to terminate each program identified in some way with the Democrats or to unfairly harass program administrators. Oversight is politicized, but that is nothing new. What is different is the perspective. Where before the advocacy context prevailed, now there is more of a tendency to reject the basic premises of programs and of much government activity. That could well change as Republicans reshape programs to their liking (welfare reform is a good example) and undo those they dislike, but that will take time. What is reasonably sure is that textbook ideas about objective evaluation will not suddenly dominate practice. Oversight will continue to have a point of view.
In 1993 Congress passed the Government Performance and Results Act (GPRA) that many hope will have a great impact on the nature of oversight. GPRA requires agencies to set goals, measure performance, and report on results. Agencies must develop five-year (strategic) plans and annual performance plans, and report each year on performance. They are to do this in consultation with Congress and others.
This is not the place (and not yet the time) to attempt a full-scale evaluation of GPRA. The act has clearly succeeded in many ways. Certainly agencies issued performance plans and GAO has worked hard to evaluate progress to date and to suggest ways to improve implementation. However, some of the problems encountered so far are rather telling. GAO reports that at only a fourth of the agencies did 50 percent or more of the managers say that they used performance information in setting priorities or allocating resources.8 Congress continues to budget using its own system, thereby making it difficult to link performance data to congressional budget decisions. Reliable performance data is difficult to obtain, particularly from the numerous bodies outside the federal government that actually implement many federal programs. Determining the links between what the federal government does and outcomes also turns out to be quite difficult because of the many outside factors that influence whether or not goals are achieved.
It is possible that GPRA will ultimately have a huge impact on the quality of oversight. But it will take a lot for this to happen. Congress will have to work with agencies in setting goals (or sign on to those they are presented with), stop micro-managing, link performance results to funds authorized and appropriated, pay careful attention to evaluations even when the results are politically difficult, and generally comport itself in ways not currently common. The chances for success will be best in a time when there is significant consensus about government policies and unified government, both conditions we often lack, but even then Congress will continue to be a political body and it is likely that its members' political needs will more often than not dominate its behavior.
My guess is that Mr. Hamilton's laudable goals for successful oversight--oversight that is comprehensive, systematic, constructive, not reflective of "petty political motives," and the like--are part of an endless quest. Because the oversight that is done will usually have a point of view, observers (and many members of Congress) will continue to complain about its quality. Reformers will continue to make suggestions. And the bureaucracy will continue to be closely watched by people who consider efficiency and effectiveness, but will have much more on their minds than that.
1. Panel Discussions before the House Select Committee on Committees, 93rd Congress, 1st Sess., Vol. 2, Part 1, pp. 15-16.
2. Lee H. Hamilton, "Oversight: A Key Congressional Function," remarks to the Congressional Oversight Workshop, 28 June 1999, http//wwics.si.edu/congress/background/oversight-hamilton.htm.)
3. Joel D. Aberbach, Keeping a Watchful Eye: The Politics of Congressional Oversight (Brookings, 1990), 200
4. House Report 105-836, p. 37.
5. Ibid., 83-84.
6. Ibid., 87.
7. Ibid., 11.
8. Government Accounting Office 01-872 T, p. 3.
