¹ Macmahon, Arthur W., “Congressional Oversight of Administration: The Power of the Purse,” Political Science Quarterly, Vol. 58 (06 and September, 1943), pp. 161–190; 380–414CrossRefGoogle Scholar. See also Schilling, Warner, “The Politics of National Defense: Fiscal 1950,” in Schilling, et al. , Strategy, Politics and Defense Budgets (New York, Columbia University Press, 1962), pp. 1–266Google Scholar; Banfield, Edward, “Congress and the Budget: A Planner's Criticism,” this Review, Vol. 43 (12, 1949), pp. 1217–1228Google Scholar; Huzar, Elias, The Purse and the Sword (Ithaca, Cornell University Press, 1950)Google Scholar; Scher, Seymour, “Conditions for Legislative Control,” Journal of Politics, Vol. 25 (08, 1963), pp. 526–551CrossRefGoogle Scholar; Wildavsky, Aaron, The Politics of the Budgetary Process (Boston, 1964)Google Scholar; Fenno, Richard, “The House Appropriations Committee,” this Review, Vol. 56 (06, 1962), pp. 310–324Google Scholar.

² Calendar years 1948 to 1962. A given budget (or fiscal) year (e.g., 1963) begins for the subcommittee with the opening of hearings in the early months of calendar 1962. Dates in this study are in terms of budget years.

³ For a member's comment to this effect, see House Appropriations Subcommittee on the Departments of Labor and Health, Education and Welfare, Departments of Labor and Health, Education and Welfare Appropriations for 1959: Hearings, 85th Cong., 2d sess. (Washington, G.P.O.), p. 201Google Scholar.

⁴ Wildavsky, op. cit., p. 87.

⁵ See Matthews, Donald R., U.S. Senators and Their World, (New York, 1960), pp. 276–78Google Scholar.

⁶ On the basis of pre-congressional occupations, the members of the Fogarty subcommittee appear somewhat less able, on the average, to deal with complex technical and political matters than other members of the Appropriations Committee. During the 1949–63 period, only 70 per cent of the subcommittee members came from occupations that would suggest a high level of relevant skills (i.e., law and other professions, journalism and politics), while 81 per cent of all Committee members came from these fields. The difference in skill-potential disappears, however, if long experience in Congress or on the Appropriations Committee can make up for an unimpressive occupational background. Subcommittee members have had more experience in both areas: 78 per cent of them in the 1949–63 period entered Congress before 1950, while only 63 per cent of all Appropriations Committee members entered so early; and 60 per cent of the subcommittee members joined the Appropriations Committee before 1950, in comparison to only 51 per cent of the entire committee.

During the period of the study, subcomittee members voted in accord with the New Republic's conception of domestic liberalism 48 per cent of the time, while the full committee voted that way 42 per cent of the time.

⁷ In the last year of the study's time span, the agencies' requests were: OE, $598 million; CB, $80 million; FDA, $28 million; HU, $13 million. Over the period 1951–63, the percentage increases in budget requests were: OE, 2120; FDA, 565; CB, 335; HU, 280. A comparative study of the agencies' budget strategies vis à vis the subcommittee revealed that OE was generally the most aggressive toward the subcommittee, HU was generally least aggressive, while CB and FDA fell in the middle range. See Sharkansky, Ira, “Four Agencies and An Appropriations Subcommittee: A Comparative Study of Budget Strategies,” Midwest Journal of Political ScienceGoogle Scholar (forthcoming).

⁸ See Berelson, Bernard, Content Analysis in Communications Research (Glencoe, 1952)Google Scholar; North, Robert C., Content Analysis (Evanston, Northwestern University Press, 1963)Google Scholar; Lasswell, Harold, ed., Language of Politics (New York, George Stewart Press, 1949)Google Scholar; and Poole, Ithiel de Sola, ed., Trends in Content Analysis (Urbana, University of Illinois Press, 1959)Google Scholar.

⁹ For a list of the categories used, see Sharkansky, Ira, “Pour Agencies and An Appropriations Subcommittee: A Comparative Study of Budget Relations,” unpublished Ph.D dissertation, University of Wisconsin, 1964Google Scholar, ch. 3.

¹⁰ This study defines an item as “covered” if it received mention in 5 per cent of the questions directed at the agency over the odd years 1949–63, or in at least 10 per cent of the questions in one year's hearings. Admittedly, “coverage” of an activity is a rough measure of thoroughness. It does not mean that the subcommittee members give their attention to every issue and ramification related to an item.

¹¹ See Alexander L. George, “Quantitative and Qualitative Approaches to Content Analysis,” in Ithiel de Sola Poole, ed., op. cit., pp. 7–32.

¹² Agency officials as well as subcommittee members perceive these as cues that will elicit Congressional questioning. Agency budget officers write the opening statements for the hearings, knowing that these are the final communications legislators will receive before questioning agency witnesses. And they know that large items, or items showing a marked change from the previous budget tend to attract attention. One of the considerations in requesting funds for an activity is, “How will it look?” If a request might appear out of place, it may be postponed in part, or divided into smaller pieces and distributed among inconspicuous budget headings. See Wildavsky, op. cit., ch. 3.

¹³ This study considers an item to be emphasized in the opening hearing statement of an agency if it is mentioned in at least 10 per cent of the statement's lines. An item is of “large absolute magnitude” if it is at least 20 per cent of the agency's total request. An item has shown “significant change” if it has shown an absolute increase or decrease of $200,000 since the last budget, or a percentage change amounting to 10 per cent of the total change in the agency's budget.

¹⁴ For example, suppose that the items emphasized in the budgets and statements of each of two agencies received mention in a mean 10 per cent of the subcommittee's questions, and that in the case of one agency, all its activities—both those emphasized and unemphasized—received mention in a mean 5 per cent of the questions. The index of subcommittee non-reliance would then be 10 minus 5, or 5. If in the second agency's case all activities received mention in a mean 8 per cent of the subcommittee's questions, the index of subcommittee non-reliance would be 10 minus 8, or 2. The congressmen would be showing less reliance on the cues of the second agency. The data for this analysis came from the hearings of 1951, 1957 and 1963, chosen because they fall near the beginning, middle and end of the period studied.

¹⁵ Because the Department Secretary and agency officials share an identity with the Administration, it might be said that the Secretary and his staff are not sufficiently independent of the agencies to be considered “non-agency” sources of information. Yet there are significant budget disputes between agency and departmental personnel. Department officials report in formal hearings and in interviews that agencies typically demand more funds than the department can allow them. From the agency view, the department is a brake against desirable expansion. Over the 1959—63 period, the department budget office reduced the requests of our sample agencies by $290.8 million, or 11 per cent of the agencies' original requests. In the hearings, departmental witnesses have been critical of agency operations. In contrast, none of the interest groups or non-subcommittee congressmen testifyng in the odd years 1951–63 criticized agency operations. They either supported agency requests, or urged the replacement of budget cuts made by the department Budget Office or the Bureau of the Budget. The data for the analysis of subcommittee questioning of non-agency witnesses came from the hearings of 1961 and 1963, chosen because it has only been in recent years that the legislators have paid significant attention to non-agency witnesses.

¹⁶ It was only in these years that information about the agencies' original budget request to the Department Budget Office was included in the hearings record.

¹⁷ These years were chosen because they fall near the beginning, middle and end of the research span.

¹⁸ Sharkansky, “Four Agencies and an Appropriations Subcommittee: A Comparative Study of Budget Strategies,” op. cit.

¹⁹ Ibid.