Hostname: page-component-586b7cd67f-2plfb Total loading time: 0 Render date: 2024-11-23T19:59:13.591Z Has data issue: false hasContentIssue false
  • English
  • Français

Where's the Party?

Published online by Cambridge University Press:  03 March 2009

Keith Krehbiel
Get access Rights & Permissions [Opens in a new window]

Abstract

Political parties are prominent in legislative politics and legislative research. Using data from the 99th Congress, this article assesses the degree to which significant party behaviour – defined and operationalized as behaviour that is independent of preferences – occurs in two key stages of legislative organization: the formation of standing committees and the appointment of conferees. Four hypotheses are developed and tested. When controlling for preferences and other hypothesized effects, positive and significant party effects are rare. A discussion addresses some criticisms of this unorthodox approach and attempts to reconcile some differences between these and previous findings.

Type
Articles
Information
British Journal of Political Science , Volume 23 , Issue 2 , April 1993 , pp. 235 - 266
Copyright
Copyright © Cambridge University Press 1993

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Rohde, David W., Parties and Leaders in the Postreform House (Chicago: University of Chicago Press, 1991), p. 2.CrossRefGoogle Scholar

2 Collie, Melissa P., ‘New Directions in Congressional Research’, Legislative Studies Section Newsletter, 10(1986), 90–2.Google Scholar

3 Aldrich, John H., ‘Modeling the Party-in-the-Legislature’ (Duke University: manuscript, 1988)Google Scholar; Aldrich, John H. ‘An Institutional Theory of a Legislature with Two Parties and a Committee System’ (Duke University: manuscript, 1990)Google Scholar; Cox, Gary W. and McCubbins, Mathew D., Legislative Leviathan: Party Government in the House (University of California at San Diego: manuscript, 1991)Google Scholar; Rohde, David W., ‘Variations in Partisanship in the House of Representatives, 1953–1988: Southern Democrats, Realignment and Agenda Change’ (Michigan State University: manuscript, 1988)Google Scholar; ‘Democratic Party Leadership, Agenda Control and the Resurgence of Partisanship in the House’ (Michigan State University: manuscript, 1989)Google Scholar; ‘Agenda Change and Partisan Resurgence in the House of Representatives’ (Michigan State University: manuscript, 1990)Google Scholar; and Parties and Leaders;, Sinclair, Barbara, The Transformation of the US Senate (Baltimore: Johns Hopkins Press, 1989).Google Scholar

4 Rohde, , ‘Agenda Change’, p. 33.Google Scholar

5 Bach, Stanley and Smith, Steven S., Managing Uncertainty in the House of Representatives: Adaptation and Innovation in Special Rules (Washington, DC: The Brookings Institution, 1989).Google Scholar

6 Brady, David W., Congressional Voting in a Partisan Era (Lawrence: University Press of Kansas, 1973)Google Scholar; Jones, Charles O., ‘Joseph G. Cannon and Howard W. Smith: An Essay on the Limits of Leadership in the House of Representatives’, Journal of Politics, 30 (1968), 617–46CrossRefGoogle Scholar; Lowi, Theodore J., ‘Party, Policy, and Constitution in America’ in Chambers, William Nisbet and Burnham, Walter Dean, eds, The American Party Systems (New York: Oxford University Press, 1967) pp. 238–76Google Scholar; Patterson, Samuel C., ‘Legislative Leadership and Political Behaviour’, Public Opinion Quarterly, 27 (1963), 399–410CrossRefGoogle Scholar; Peabody, Robert L., ‘Party Leadership Change in the United States House of Representatives’, American Political Science Review, 61 (1967), 675–93CrossRefGoogle Scholar; Ripley, Randall B., ‘The Party Whip Organization in the US House of Representatives’, American Political Science Review, 58 (1964), 561–76CrossRefGoogle Scholar; and Party Leaders in the House of Representatives (Washington, DC: The Brookings Institution, 1967)Google Scholar; and Sinclair, Barbara, Majority Leadership in the US House (Baltimore: Johns Hopkins Press, 1983)Google Scholar; and The Transformation of the US Senate.

7 Patterson, Samuel C. and Caldeira, Gregory A., ‘Party Voting in the United States Congress’, British Journal of Political Science, 18 (1988), 111–31, p. 111.CrossRefGoogle Scholar

8 Collie, Melissa P., ‘Electoral Patterns and Voting Alignments in the US House, 1886–1986’, Legislative Studies Quarterly, 14 (1989), 107–107CrossRefGoogle Scholar; Collie, Melissa P. and Brady, David W., ‘The Decline of Partisan Voting Coalitions in the House of Representatives’, in Dodd, Lawrence C. and Oppenheimer, Bruce I., eds, Congress Reconsidered, 3rd edn (Washington, DC: Congressional Quarterly Press, 1985) pp. 272–87Google Scholar; Cox, and McCubbins, , Legislative LeviathanGoogle Scholar; Hurley, Patricia A. and Wilson, Rick K., ‘Partisan Voting Patterns in the US Senate, 1977–86’, Legislative Studies Quarterly, 14 (1989), 225–50CrossRefGoogle Scholar; Rohde, , Parties and LeadersGoogle Scholar; Sinclair, Barbara, ‘From Party Voting to Regional Fragmentation, 1933–1956’, American Politics Quarterly, 6 (1978), 125–125CrossRefGoogle Scholar; and Truman, David B., The Governmental Process (New York: Alfred A. Knopf, 1951)Google Scholar; and The Congressional Party (New York: Wiley, 1954).Google Scholar

9 For an extreme but exceptionally lucid example of this phenomenon, see Hedrick Smith's discussion of the behaviour of Howard Baker, Strom Thurmond and other Senate Republicans in 1981 when confronted with the necessity, as a new majority of the president's party, of raising the debt ceiling. See Smith, Hedrick, The Power Game (New York: Random House, 1988), pp. 459–60.Google Scholar

10 Party objectives could be defined, for example, as the party's median member's preferred policy in a given unidimensional jurisdiction.

11 See, for example, Fenno, Richard F. Jr, The Power of the Purse: Appropriations Politics in Congress (Boston, Mass.: Little, Brown, 1966)Google Scholar; Manley, John F., The Politics of Finance: The House Committee on Ways and Means (Boston, Mass.: Little, Brown, 1970)Google Scholar; Price, David E., Who Makes the Laws? Creativity and Power in Senate Committees (Cambridge: Schenkman, 1972)Google Scholar; Smith, Steven S. and Deering, Christopher J., Committees in Congress (Washington, DC: Congressional Quarterly Press, 1984)Google Scholar; Eulau, Heinz, ‘Committee Selection’, in Loewenberg, Gerhard, Patterson, Samuel and Jewell, Malcom, eds, Handbook of Legislative Research (Cambridge, Mass.: Harvard University Press, 1985).Google Scholar

12 Jones, Charles O., ‘Joseph G. Cannon and Howard W. Smith: An Essay on the Limits of Leadership in the House of Representatives’, Journal of Politics, 30 (1968), 617–46.CrossRefGoogle Scholar

13 Smith, , The Power Game, p. 479.Google Scholar

14 This is not to claim that parties do not try to exert their powers on substantive outcomes, too. Nor is it to deny the relationship between procedure and substance. Rather, the argument is of the ‘to-the-extent’ form: to the extent that a line can be drawn between matters of procedure and matters of substance, the analysis focuses on the side of the line where party effects are more likely to be manifested.

15 See, for example, Masters, Nicholas A., ‘House Committee Assignments’, in Rieselbach, N. Leroy, ed, The Congressional System: Notes and Readings (Belmont, Calif.: Wadsworth, 1970)Google Scholar; and Shepsle, Kenneth A., The Giant Jigsaw Puzzle (Chicago: University of Chicago Press, 1978).Google Scholar

16 For mean-based tests, see Krehbiel, Keith, ‘Are Congressional Committees Composed of Preference Outliers?’, American Political Science Review, 84 (1990), 149–63CrossRefGoogle Scholar. For essentially the same conclusions derived using median-based non-parametric tests, see Groseclose, Timothy J., ‘Median-Based Tests of Committee Composition’ (Stanford University: manuscript, 1992).Google Scholar

17 The studies are only suggestive since the inability to reject a null hypothesis does not imply truth of the null hypothesis of no difference, rather, it signifies the lack of evidence for the alternative hypothesis of a significant difference. Nor, as Hall and Grofman suggest by their criticism of ‘categorical accounts’ of committee assignments, do these patterns of findings imply that all committees are alike (see Hall, Richard and Grofman, Bernard, ‘The Committee Assignment Process and the Conditional Nature of Committee Bias’, American Political Science Review, 84 (1990), 1149–66)CrossRefGoogle Scholar. Indeed, previous studies report considerable cross-committee variation.

18 This party-based interpretation of studies that had no party basis is necessarily speculative. For empirical studies of committee composition that include party breakdowns, see Coker, David C. and Crain, W. Mark, ‘Legislative Committees as Loyalty-Generating Institutions’ (George Mason University: manuscript, 1990)Google Scholar; Cox, and McCubbins, , Legislative LeviathanGoogle Scholar; Groseclose, , ‘Median-Based Tests’Google Scholar, and Strahan, Randall and Weaver, R. Kent, ‘Subcommittee Government and the House Ways and Means Committee’ (Emory University: manuscript, 1989).Google Scholar

19 The present study focuses exclusively on the 99th Congress (1985–86). There are no apparent reasons for believing that this Congress is atypical and some evidence that it is typical. (See Hinckley, Katherine A., ‘Party Ratios on Congressional Committees and Subcommittees, 80th–99th Congresses’ (University of Akron: manuscript, 1986)Google Scholar for findings on party ratios and Krehbiel, , ‘Preference Outliers’Google Scholar, for evidence of committee composition in other congresses.) None the less, as always, replication and generalization would be desirable.

20 The terms ‘majority-party committee stacking’ in this context refers to over-representation of majority-party policy positions. As the following examples indicate, it is not necessary to settle on any single definition of over-representation, however chamber-mean or chamber-median based definitions are intuitive and computationally convenient.

21 The term ‘high demander’ is always defined and measured with reference to a given committee and, more specifically, an issue or political ‘commodity’ within the committee's jurisdiction. See the appendix for examples and details.

22 Jurisdictionally relevant interest-group ratings of legislators are used as measures of preferences, although constituency characteristics could also be used for some committees as in, for example. Hall, and Grofman, , ‘The Committee Assignment Process’Google Scholar. See the appendix to this article for a brief discussion of this and related issues.

23 For example, using the chamber median as the cutpoint between low and high demanders according to NFU ratings, the resulting denominators on which p2 and p3 are based are only 2 and 42, respectively.

24 The data and committee-rating pairs are the same as those used in Krehbiel, , ‘Preference Outliers’.Google Scholar

25 This specification is simple in comparison, say, to the analysis in Shepsle, 's Giant Jigsaw PuzzleGoogle Scholar. Although both analyses use individual-level data, their substantive aims are quite different. Shepsle was mainly concerned in self-selection and therefore compared (among other things) freshmen requests or non-freshmen transfer requests with intra-party assignments (Democrats only). I am more concerned with overall partisan effects than with intra-party individual decision making. As such, the econometric specification focuses on overall committee membership at a given time rather than first-time assignments. These approaches are not radically different from, or inconsistent with, one another. Indeed, this equation might be viewed as a reduced-form (albeit bipartisan) version of Shepsle's.

26 Equation (1) is an econometric equivalent of Pr(COMMTTTEE = 1) = Φ(γ1DEM + γ2DEM × PREFERENCE + γ3REP + γ4REP × PREFERENCE), where the party dummy variables have the obvious definitions. While this equation is perhaps more intuitive for comparing parties, the β coefficients in Equation 1 are somewhat easier to interpret because they parallel the three substantive hypotheses and highlight the present majority-party focus.

27 A critical t of 1.65 is used, indicating significance at the 0.05 level in a one-tailed test. See the appendix for the complete estimates, including the t-statistics which, of course, enable applications of whatever standard of significance is desired.

28 Murphy, James T., ‘Political Parties and the Porkbarrel: Party Conflict and Cooperation in House Public Works Committee Decision Making’, American Political Science Review, 68 (1974), 169–85.CrossRefGoogle Scholar

29 Let denote the mean NFU rating, sx its standard deviation, and the estimates reported in the appendix. Then the cell entries are calculated as follows: low-demand Republican, Φ( + ( – sx)); medium-demand Republican, Φ( + ); high-demand Republican, Φ( + ( + sx)); low-demand Democrat, Φ( + + ( – sx) + ( – sx)); medium-demand Democrat, Φ( + + + ); high-demand Democrat, Φ( + + ( + sx) + ( + sx)).

30 Furthermore, the PARTY coefficient in Equation 2 is negative and significant. A likelihood ratio test, however, suggests that Equation 2 is not as good a fit as Equation 1 for this committee-rating pair.

31 While these are instances in which, a priori, it is not clear what a high demander is, the appendix offers guidelines that allow the reader to choose his or her preferred interpretation.

32 According to Krehbiel, in ‘Preference Outliers’Google Scholar and Groseclose, in ‘Median-Based Tests’Google Scholar, only Armed Services is a preference outlier. Hall, and Grofman, (‘The Committee Assignment Process’)Google Scholar would probably claim that Agriculture is too (which it nearly was in Krehbiel's analysis but not in Groseclose's), though their analysis pertained only to the Senate Agriculture Committee.

33 Shepsle, , Giant Jigsaw Puzzle.Google Scholar

34 Committee and chamber means are 51.9 and 52.4 respectively; medians are 53 and 50, respectively.

35 None of the committees analysed were ‘exclusive’ as defined by the House because it is difficult or impossible to find satisfactory jurisdiction-specific ratings for Appropriations, Budget, Rules and Ways and Means. As a less satisfactory substitute, I nevertheless estimated the same equations for these committees using ADA ratings. While ADA ratings are very broad in terms of the votes that figure into their calculation, they tend to be correlated strongly with other measures used above. As such, the findings were stunning in their lack of support for all hypotheses. In the eight equations estimated – two for each committee – not a single coefficient was significant (except for the constant term).

36 On the importance of conference committees in the legislative process, see Fenno, , The Power of the PurseGoogle Scholar; McCown, Ada C., The Congressional Conference Committee (New York: Columbia University Press, 1927)Google Scholar; Longley, Lawrence D. and Oleszek, Walter J., Bicameral Politics: Conference Committees in Congress (New Haven, Conn.: Yale University Press, 1989)Google Scholar; Smith, Steven S., Call to Order: Floor Politics in the House and Senate (Washington, DC: The Brookings Institution, 1989)Google Scholar; Steiner, Gilbert Y., The Congressional Conference Committee (Urbana: University of Illinois Press, 1951)Google Scholar; Strom, Gerald S. and Rundquist, Barry S., ‘A Revised Theory of Winning in House-Senate Conferences’, American Political Science Review, 71 (1977), 448–448CrossRefGoogle Scholar; and Vogler, David J., The Third House.Conference Committees in the United States Congress (Evanston, Ill.: Northwestern University Press, 1971)Google Scholar. Recent formal theories, too, adopt this perspective by identifying conference committees as bastions of committee power, and as opportunities to pursue partisan advantage, given the speaker's unchallengeable right to appoint conferees. See, respectively, Shepsle, Kenneth A. and Weingast, Barry R., ‘The Institutional Foundations of Committee Power’, American Political Science Review, 81 (1987), 85–85CrossRefGoogle Scholar; and Nagler, Jonathan, ‘Strategic Implications of Conference Selection in the House of Representatives’, American Politics Quarterly, 17 (1989), 54–79CrossRefGoogle Scholar. The relationship between this literature and party effects on conferee selection is less clear, however. For example, in a recent review of Longley, and Olcszek, 's book, Bicameral PoliticsGoogle Scholar, Patterson writes: ‘Longley and Oleszek's heavy reliance on the extant literature about Congress sometimes leads them astray. This appears strikingly in their assertions about the weakening of party leadership in Congress, especially the House. In contrast to the authors' claims, I believe careful research would show that House leaders carry much more influence in conference outcomes in the 1980s than did Speaker Rayburn in the 1950s … At a minimum, their argument points to the need for much more substantial research on the influences of party and committee leaders on conference decision making … Today congressional conference committees are open for systematic analysis … It is up to other scholars to take up these cudgels' (American Political Science Review, 84 (1990), 661–3).Google Scholar

37 Shepsle, and Weingast, , ‘Institutional Foundations’.Google Scholar

38 Krehbiel, Keith, Information and Legislative Organization (Ann Arbor: University of Michigan Press, 1991).CrossRefGoogle Scholar

39 Lists of the bills can be found in Krehbiel, , Information and Legislative Organization, pp. 225–8.Google Scholar

40 The appendix provides additional information on measures. See Krehbiel, , Information and Legislative Organization, chap. 6Google Scholar, for a lengthy discussion of distributive and informational hypotheses and findings on postfloor politics, though one that does not consider party effects.

41 The positive OUTLIER effect cannot be interpreted as support for the hypothesis of gains from trade among high demanders, however. The measure of outliers is the absolute deviation in COPE ratings from the House mean – not simply high demand for labour benefits. As such, high and low demanders for pro-labour services have a probabilistic advantage.

42 For an excellent statement of two competing conceptualizations of the role of parties in political systems more broadly, see Plotnick, Robert D. and Winters, Richard F., ‘Party, Political Liberalism, and Redistribution: An Application to the American States’, American Politics Quarterly, 18 (1990), 430–58.CrossRefGoogle Scholar

43 When theorizing about preference formation, though, it is essential to draw a sharp distinction between preferences over policies (which are the objects of legislative choice, e.g., bills, amendments) and preferences over outcomes (which are the consequences of the implementation of policies, for example, the well-being of constituents). When political scientists casually refer to the formation of preferences, they almost invariably (but implicitly) have in mind preferences over policies – not preferences over outcomes. Preferences over outcomes tend to be stable and thus are reasonably taken as given in most formal theories. Preferences over policies are considerably less stable because of their sensitivity to information that decision makers acquire about the relationship between, for instance, an amendment and its consequences. Almost surely, a key role of parties in legislatures is informational in this sense, and, thus, a related branch of game theory can potentially accommodate notions of preference formation over policies. However, a well-developed theory of this form does not yet exist.

44 Likewise, we can imagine a theory in which preferences precede partisanship. Although this view may have substantial appeal in the electoral arena, its appeal diminishes precipitously with respect to the sorts of intra-legislative behaviour on which this study focuses, since members join the Congress with partisan affiliations.

45 The parallel finding for the Interior Committee is of the expected sign and significant. However, for this committee-rating pair, the direct party effect is significant and of the wrong sign.

46 Schattschneider, E. E., Toward a More Responsible Two-Party System’, American Political Science Review, 44 (1950), SupplementGoogle Scholar; Epstein, Leon D., ‘What Happened to the British Party Model?’, American Political Science Review, 74 (1980), 9–22.CrossRefGoogle Scholar

47 Schwartz, John E., ‘Exploring a New Role in Policy Making: The British House of Commons in the 1970s’, American Political Science Review, 74 (1980), 23–37CrossRefGoogle Scholar, see especially Tables 1–4.

48 Schwartz also occasionally attributes some of these changes to the demise of ‘the parliamentary rule’ that ‘government backbenchers in the Commons would [not] cross-vote to defeat their own government on the floor’ (Schwartz, , ‘Exploring a New Role’, p. 33)Google Scholar. This demise, however, is more definitional of the phenomenon of interest than it is a cause. The primary cause in the breakdown of the ‘rule’ (which is actually a regularity in behaviour rather than a genuinely binding constraint on behaviour) was the increasingly heterogeneous preferences that Schwartz identifies elsewhere in his superb study.

49 Broder, David S., The Party's Over: The Failure of Politics in America (New York: Harper & Row, 1971).Google Scholar

50 Cain, Bruce, Ferejohn, John and Fiorina, Morris, The Personal Vote (Cambridge, Mass.: Harvard University Press, 1987).CrossRefGoogle Scholar

51 Sinclair, Barbara, ‘Majority Party Leadership Strategies for Coping with the New House’, Legislative Studies Quarterly, 6 (1981), 391–414.CrossRefGoogle Scholar

52 Mayhew adds yet another, inter-branch layer to the argument, concluding that whether or not the presidency and Congress are controlled by opposite parties has little bearing on the number of investigations conducted by Congress, on the ability of Congress to pass major legislation, or on the margins by which such legislation is passed. See Mayhew, David R., Does It Make a Difference Whether Parly Control of the American National Government is Unified or Divided? (New Haven, Conn.: Yale University Press, 1991).Google Scholar

53 This, in turn, may have several sources: actual changes in individual voters' preferences, redrawn district lines, local politics, etc.

54 Rohde, , Parties and Leaders;Google ScholarSmith, and Bach, , Managing Uncertainty.Google Scholar

55 Rohde, , Parties and Leaders, p. 192.Google Scholar

56 The data are the same as those used Krehbiel, , ‘Preference Outliers’Google Scholar, and were obtained from Legi-Slate, a subsidiary of the Washington Post. Some of these measures differ mildly from those available elsewhere for two reasons. First, Legi-Slate does not count absences as ‘incorrect’ votes as do some, but not all, interest groups. Second, in instances in which the separate ratings are available for two sessions within the same Congress, ratings were averaged across the sessions.

57 For an example of a more formal statement of the underlying model in constituency characteristic-based measures of preferences, see Krehbiel, Keith, ‘Constituency Characteristics and Legislative Preferences’, Public Choice (forthcoming, 1993).Google Scholar

58 For a more detailed defence, see Krehbiel, , ‘Preference Outliers’.Google Scholar

59 Hall, and Grofman, , ‘The Committee Assignment Process’.Google Scholar

60 Groseclosc, , ‘Median-Based Tests’Google Scholar; Krehbiel, Keith, ‘Deference, Extremism and Interest Group Ratings’ (Stanford University: manuscript, 1992).Google Scholar

61 Krehbiel, , ‘Preference Outliers’.Google Scholar

62 Fenno, Richard F. Jr, Congressmen in Committees (Boston, Mass.: Little, Brown, 1973).Google Scholar