Purchasing Support or Rewarding Loyalty



Ideological Congruence Versus Electoral Success: Distribution of Party Organization Contributions in Senate Elections 1990-1998*

Timothy P. Nokken

Department of Political Science

University of Houston

Houston, TX 77204-3474

tnokken@pop.uh.edu

Abstract

One of the key ingredients for a successful Congressional campaign is money. Rarely do candidates who lack financial resources succeed in their bid for office. Over the past decade, Democratic and Republican national party organizations have taken a much more active role in congressional campaigns. A key component of such involvement is the distribution of national party funds to House and Senate candidates. I seek to determine whether parties are “single minded seekers of reelection,” or whether they seek to reward individuals who share a “correct” ideology. I use data from Senate elections from 1990 to 1998 that involve either incumbent senators or House members seeking Senate seats to determine which trait parties value more: competitiveness or ideology.

While some scholars argue that American political parties are in a state of decline, one relatively recent development is the increased activity of the national party organizations in congressional elections. National party organizations like the Democratic and Republican National Committees (DNC are RNC), Democratic and National Republican Senatorial Committees (DSCC and NRSC), the Democratic Congressional Campaign Committee (DCCC), and the National Republican Campaign Committee (NRCC) work exert tremendous amounts of energy to raise and distribute campaign funds to assist House and Senate candidates. As a result of such efforts, national party organizations are becoming increasingly important players in congressional elections. While the power of subnational party organizations have continued to decline, the national party organizations appear to be enjoying increasing power and influence.

One poignant example of the presence of the national parties’ in congressional races is the 1996 reelection bid of campaign finance reform champion Russ Feingold (D-WI).[1] Feingold and his opponent, Republican House member Mark Neumann, agreed to a self-imposed spending limit of one dollar per voter in their race, a spending ceiling of $3.8 million. That limit, however, did not deter the national party organizations from getting involved. The DSCC bought time to air a negative spot on Neumann, but subsequently pulled the ad after Feingold protested. The NRSC poured substantial amounts of money into Neumann’s campaign late in the race. Sen. Mitch McConnell’s (R-KY, and RSCC chair) late foray into the Wisconsin campaign prompted some to question publicly whether McConnell was sending resources to where they could do the most good or whether his behavior was better characterized as an effort to “get” Feingold, his Senate nemesis on the issue of campaign finance (Wilcox, 2000).

The Wisconsin race serves as a point of departure with which to investigate the national party organizations’ strategies for distributing campaign funds. While some openly questioned McConnell’s motivation for his decision to spend heavily on Neumann’s campaign at the last minute, Neumann was a competitive quality candidate who lost by a mere 35,000 votes. The chairs of the respective parties’ congressional and senatorial campaign committees contend that the strategy is not to target recipients based on their ideological predispositions but to provide funds to competitive candidates. The story from Wisconsin, however, highlights the potential tensions party organizations must consider in their work to provide funds to their candidates. Namely, leaders must determine whether to pursue a strategy in which they attempt to elect strong partisans who share the party’s preferences on policy, or one where they overlook policy preferences and contribute money and resources to those candidates who have the best chance at winning. I revisit this topic in order to determine whether individual candidates’ policy preferences play a significant role in the parties’ distribution of campaign funds. I analyze the parties’ donations to a select group of candidates for the United States Senate from 1990 to 1998 to determine whether ideological or electoral concerns best characterized the distribution of campaign funds. I take advantage of a superior measure of candidate ideology to determine its role, if any, on party contribution strategies.[2]

The role of party organizations in the funding of congressional elections generates a number of interesting lines of investigation for political scientists. Such activities have implications for the study of political parties, obviously, of the importance of money in congressional campaigns, and potentially for party unity in the legislative arena. In this section, I describe the types of expenditures party organizations make to individual congressional candidates as well as the regulations governing such transfers of funds.

Campaign Finance Laws and the Activities of Party Organizations

The Federal Election Campaign Act of 1974, and the subsequent revisions of 1979, allows both local and national party organizations to contribute to congressional campaigns. These organizations can make direct contributions whereby parties simply contribute cash directly to a campaign, or make coordinated expenditures that essentially allow them to purchase important campaign services in bulk (Herrnson, 2000, 94). The party organizations that make direct contributions and coordinated expenditures work closely with the individual campaigns when distributing such funds.[3] Starting in 1996, parties were allowed to make what are known as independent expenditures.[4] Contributions of this sort are made by the party organization either to increase support for their candidate or to attack their opponent, but must be done without input or consultation from individual campaigns.

Both types of expenditures are subject to limits defined in the FECA. Direct contributions are limited to $5,000 per candidate per election (primary, general, runoff). The limits on coordinated expenditures are determined in one of two manners: either two cents times the state’s voting age population times a cost of living adjustment, or $20,000 times the cost of living adjustment, whichever is highest. Limits on coordinate expenditures for the 2000 Senate election cycle, for instance, range from $67,560 in states with a single Representative to $1,636,438 in California. (FEC RECORD, March 2000a)

The national political parties pour millions of dollars into congressional races each election cycle. In 1998, for instance, the two major parties raised in excess of $287 million and spent nearly $8.2 million in direct contributions and over $18.68 million in coordinated expenditures. Table 1 displays aggregate party receipts and expenditures as well as documents spending by the two parties by types of races for the 1996 and 1998 election cycles. Table 2 displays total receipts, party contributions (by party), and coordinated expenditures (by party) for each election cycle from 1990 to 1998. Again, while these expenditures comprise only a small proportion of the parties’ total receipts, they are also limited by the regulations detailed in the FECA. If one includes party expenditures of soft money (primarily through party independent expenditures either for or against candidates) the parties’ contributions to congressional races reach even loftier levels.[5]

Table 1 about here

Table 2 about here

Competing Incentives: Competitiveness vs Ideology

Despite receipts of millions of dollars each election cycle, party committees are not able to distribute the maximum allotment of funds to every candidate. In short, they must choose to support some candidates at the expense of others. Parties, then, are faced with difficult choices when determining who will enjoy the fruits of their fundraising prowess and who will go without. The parties must determine what they value most in a candidate: ideological congruence or electoral appeal

The conventional view regarding allocations of party campaign funds is that they go to where they do the most good: to endangered incumbents, competitive challengers, and hotly contested open seats. When asked about his strategy, RSCC Chair McConnell stated “…we make the decisions every day by the numbers. I get tracking polls from the states where we’re competitive across the country. I have no particular bias in favor or against any of our candidates. The only issue is, How close are you?” (Herrnson, 2000, 92). The conventional wisdom is compelling. It is hard to believe that the parties will focus their resources on candidates who may share their policy preferences but have no chance of winning. That said, it certainly seems plausible that the parties take the policy preferences of their candidates into consideration when deciding how to distribute funds. I seek to determine, after controlling for a number of circumstances including competitiveness, whether party resources will be more likely to flow to those individuals the party can be fairly certain share a Democratic or Republican view on the matters facing Congress.

Senate Elections and Party Contributions

The importance of campaign spending has been a hotly debated topic for several years (see, for instance, Jacobson, 1997; Green and Krasno, 1988; 1990). Relatively few, however, have studied party expenditures in congressional elections (see, however, Bianco, 1999; Damore and Hansford, 1999; Cantor and Hernson, 1997; Jacobson, 1993; Leyden and Borrelli, 1990; Shively and Wilson, 1994). Generally, scholars are interested in determining whether donations serve to reward loyalty, generate party loyalty, or simply to enhance the prospects of electoral victory without regard for ideology. With few exceptions, they find that party resources are aimed at competitive races rather than toward ideologically appealing candidates.[6] Such studies, however, focus on House races and primarily on incumbents (although Damore and Hansford (1999) expand their dataset to include House challengers). Focusing solely on House races makes it difficult to test the importance of ideological factors because there is not a common metric with which to compare House members and non-members which makes it difficult to get a true sense of how ideological concerns color party strategy independent of incumbency. As a result, the focus House elections provides one with a potentially skewed picture of American congressional elections. In an effort to address some of these problems, I revisit the topic of party contributions by analyzing party donations to a specific set of Senate candidates – incumbent Senators seeking reelection and former House members seeking Senate seats in general election contests.[7]

The drastically different institutional designs of the House and Senate provide one obvious rationale for expanding studies to include Senate elections. Different terms of office, and different constituencies, for instance, suggest one might observe distinct contrasts between House and Senate elections. One such contrast is the that Senate elections are more competitive than House elections, that Senate challengers are successful more often than House challengers.[8] Some of the structural differences between House and Senate races provide scholars with analytical leverage with which to test competing hypotheses about how parties determine who will and who will not receive limited campaign resources.

One possible implication of such institutional differences is the level of electoral competition for Senate seats (see Krasno, 1991). The Senate is viewed as a highly attractive office for ambitious House members, which means a number of Representatives leave the House to seek Senate seats (Black, 1972; Rohde, 1979; Schlessinger, 1966).[9] As is the case with House elections, quality challengers are more likely to unseat incumbents than those challengers who have never held elective office. The statewide constituencies of Senate seats result in important decisions regarding allocations of campaign resources, like capitalizing on economies of scale in some campaign activies like television media buys, for instance.

The prevalence of House members leaving to seek Senate seats not only serves to make Senate races more competitive, but it provides scholars with leverage on an important question regarding the importance of ideology in the distribution of party campaign funds. Measuring the ideology of most challengers is difficult for two reasons. First, it is difficult to find any sort measures of preferences for most of the people who seek office, even if they have served in an elective office. Secondly, in those instances in which measures do exist, they are not comparable to those measures available for the incumbent candidates.[10] For example, an ADA score for a US Senator is not comparable to an ADA score for a member of the California State Assembly. Poole’s (1998) common space NOMINATE scores overcome the problem of comparability, utilizing a scaling procedure to place Senators and Representatives in the same policy space. These common space NOMINATE scores allow for direct comparisons of House and Senate members both across time and across institutions. As a departure from other analyses, I limit my study to Senate elections, further winnowing the data set to include incumbent senators and former House members (and one former senator who sought to recapture the seat he lost six years ealier) to take advantage of the common space NOMINATE scores to gain insight on possible party committee disbursements.

Data and Methods

The data set is comprised of incumbent senators seeking reelection, former House members seeking Senate seats, and one former senator seeking to recapture the seat he lost from 1990 to 1998, for a total of 183 candidates (95 Republicans and 88 Democrats). Incumbents comprised 134 of the candidates; challengers, 49 candidates; and 30 candidates in open seats. I use this set of members to test a series of hypotheses regarding party strategies for disbursing limited campaign resources. Parties have competing incentives in the distribution of funds. Ideally, all money would have a significant and beneficial effect that would allow them contribute to those candidates they liked best. Presumably, the candidates they would like best are the ideological congruent candidates, those who share the party’s preferences over policy alternatives. If, however, parties are vote maximizers as Downs (1957) characterizes them to be, one would expect the parties to shift resources to those races in which they will do the most good in order to maximize the party’s probability of winning the seat with little concern for ideological congruence. Finally, one might expect parties to give primarily to known quantities. In other words, the parties would choose to give more funds to incumbents. There exists some evidence that Democrats pursued a strategy that favored safe incumbents in the early 1990s (Jacobson, 1993; Shively and Wilson, 1994).

Previous studies provide the basis for a set of hypotheses regarding party strategies for disbursing campaign funds. I test the following three hypotheses:

1) Ideological Congruence Hypothesis: Campaign funds will flow to individuals who most resemble the party’s policy preferences.

2) Vote Maximizing Hypothesis: Parties are agnostic with respect to ideology and distribute funds to competitive races where they can do the most good.

3) Incumbency Bias Hypothesis: Parties may prefer to give to known quantities. If so, one should expect party committees to give incumbents more money.

Dependent Variable

The dependent variables measure how much the national party committees contribute to individuals seeking Senate seats. The data used in this study are found at the ftp site at the Federal Election Commission’s web site.[11] I use the Report on Financial Activity for Senate and House Campaigns files for each election cycle from 1990 to 1998, to identify how much money the political parties contribute to the respective Senate candidates in the form of major party contributions, coordinated expenditures, and independent expenditures (both for and against).[12] It is important to distinguish between those contributions subject to explicit limits and those that are not. The first variable is a measure of the amount of restricted donations (major party contributions and party coordinated expenditures) the parties made to candidates. I sum the direct contributions and coordinated expenditures and divide that sum by the FECs limit on coordinated expenditures creating a percentage of the maximum party contribution allowed by law. Using a ratio as the dependent variable for direct comparisons of party activity over time without the need to adjust dollar amounts for inflation. Coding the dependent variable in this manner is also important because the spending limits on coordinated expenditure vary across states depending on size. Seemingly small contributions in some states might represent a larger proportion of the maximum allowable contribution than much larger donations in states with larger populations. The second dependent variable is simply the amount of money the parties spent on independent expenditures, both for and against, and are measured in 1998 dollars.[13]

Independent Variables

A key component of this research centers on whether ideology plays a role in the distribution of campaign funds. In response to past studies, I constructed the data set to ensure that each candidate has an ideology score directly comparable to other candidates in the data set. I use use Poole’s common space NOMINATE scores to measure ideology. These scores range from –1 for the most liberal members and +1 for the most conservative. Common space scores allow for direct comparison of Senate and House members, thereby enabling me to draw stronger conclusions about the importance of ideology in party disbursements. I utilize the common space NOMINATE scores to construct two measures. The first is simply a candidate’s common space NOMINATE score, used to determine whether funds are given to more liberal (conservative) candidates. As a more precise measure of ideological congruence, I construct a DISTANCE variable, which is simply the absolute difference between a candidate’s NOMINATE score and the corresponding Senate party median NOMIANTE score for the Congress in which the election takes place (for instance, 1994 Senate candidates’ common space NOMINATE score would be compared to their parties’ Senate median for the 103rd Congress). If parties distribute funds based on ideology, one would expect to find significant coefficients on the NOMINATE variable, and more precisely, statistically significant and negative coefficients on the DISTANCE variable – parties are less likely to give funds to candidates the further they are from the party median.

By including House members in my dataset, I am at least partially controlling for challenger quality because every candidate in the data set is a quality candidate. Simply put, House member, on average, prove to be formidable challengers for incumbents (and open seats). Additionally, a number of incumbents are viewed as vulnerable which may (or may not) lead to spirited challengers, intense campaigns, and the possibility of an influx of dollars from the political parties in an effort to protect the incumbent or assist strong challengers. In February of each election year, Congressional Quarterly Weekly Report handicaps each party’s chances of retaining Senate seats, rating states as highly vulnerable, vulnerable, potentially vulnerable, and probably secure. These categories were used to code a THREAT variable. Highly vulnerable states were coded two, vulnerable states coded one, and the rest were coded zero. Coding was identical for incumbents, challengers, and candidates in open seats. If, as the vote maximizer hypothesis suggests, parties focus their resources on competitive races, highly vulnerable and vulnerable states would appear to be the most competitive, hence, a positive and statistically significant coefficient on the THREAT variable.

To test the incumbency bias hypotheses, I accounted for incumbency status with a dummy variable coded one for incumbents, zero otherwise. If incumbents are more likely to receive party financing, one would expect to find a positive and statistically significant coefficient on this variable. Other factors influence election outcomes. For instance, open seat races are thought to be more competitive on average than races matching an incumbent against a challenger. I include a dummy variable coded one for those candidates running for an open seat, zero otherwise. I include a dummy variable for party, coded one for Democrats, zero for Republicans. Finally, I include but do not report yearly dummy variables for elections held from 1990 to 1996.

RESULTS

I estimated tobit[14] models using both dependent variables. I ran the model using both the common space NOMINATE scores and the distance measure on the entire dataset and then separately for each party. The results (displayed in Table 3) provide no evidence to suggest that party campaign contributions are distributed contingent on any sort of ideological purity test. Overall, Democrats received less from their party than did Republicans. The only statistically significant ideology-related variable is the distance variable for Republican candidates, but the sign is in the wrong direction (the further away the candidate from the Republican Senate median, the higher the percentage of funds). Among Democrats, the sign of the coefficient on the Distance variable is negative suggesting candidates closer to the Democratic Senate median receive more funding, but the variable does not achieve statistical significance. Likewise, there is no evidence to support the incumbency bias hypothesis. Parties contribute fewer dollars to the campaigns of incumbents, which is not wholly unexpected given their ability to use incumbency as a resource in fundraising.

Table 3 About Here

The Ideological Congruence Hypothesis is easily rejected. Likewise, the results lend credence to the Vote Maximizing Hypothesis. In each model, the coefficient on the threat variable is positive and highly significant. Supporting the anecdotal evidence about the factors that determine party strategy for distributing funds, money flows to the most competitive races where it does the most good. Those candidates seeking Senate seats categorized as “highly vulnerable” and “vulnerable” receive a higher proportion of funds from the political parties. The incumbent’s party, no doubt, is contributing in an effort to hold onto the seat, while the party of the challenger smells blood and feels an influx of party dollars can increase the likelihood of capturing an additional Senate seat.

I use the same specifications to estimate a tobit model to predict the total amount spent by the parties on independent expenditures (both for and against) from the 1996 and 1998 elections. The results (displayed in Table 4) are not nearly as conclusive as those from the previous models.[15] NOMINATE scores are a significant determinant of expenditures in favor of a candidate, but not for independent expenditures against. The distance variable has no statistically significant effect on the distribution of such funds. The Threat variable is statistically significant for party contributions in favor of candidates (statistically significant at the 0.05 level in the model using NOMINATE scores and at the 0.10 level in the model using the distance measure). Vulnerable and highly vulnerable status results in a significantly larger influx of funds from the national parties. Once again, incumbency has a negative effect on party contributions, with incumbents receiving less help from their parties (the coefficients are negative in each model, and statistically significant in the model predicting independent expenditures against).

Table 4 about here

Conclusions

The goal of this paper was to supplement and extend the research on the distribution of party committee funds to legislative campaigns. The contributions of the paper are twofold. First, and most obviously, the work focuses on Senate campaigns whereas most of the existing works analyze House elections. Secondly, I utilize a measure of ideology that allows for the direct comparison of both incumbent senators and former House members seeking Senate seats. One of the limitations in the analyses focusing on House elections is the lack of a metric with which to compare directly the policy preferences of challengers and incumbents. The findings in the paper supplement what a number of the studies of House elections determined, that ideology and incumbency are less important in the distribution of party committee funds. In the end, the parties are true Downsian vote maxmizers, distributing monies to the most competitive races hoping to hold onto the threatened seats they already hold, or to capture those seats in states where the incumbent party may be vulnerable.

Works Cited

Bianco, William T. 1999. “Party Campaign Committees and the Distribution of Tally Program Funds.” Legislative Studies Quarterly 24:451-469.

Black, Gordon S. 1974. “A Theory of Political Ambition: Career Choices and the Role of Structural Incentives.” American Political Science Review 66:144-159

Cantor, David M. and Paul S. Herrnson. 1997. “Party Campaign Activity and Party Unity in the House of Representatives.” Legislative Studies Quarterly 22:393-415.

Clymer, Adam. 2000. “House Democrats Have Money Edge on GOP” New York Times. June 11, 2000. .

Damore, David F. and Thomas G. Hansford. 1999. “The Allocation of Party Controlled Campaign Funds in the House of Representatives.” Political Research Quarterly 52:371-385.

Downs, Anthony. 1957. An Economic Theory of Democracy New York:Harper.

Federal Election Commission. 1996. FEC Campaign Guide for Political Party Committees

Federal Election Commission. 2000a. The Record Volume 26 (March).

________. 2000b. “Federal Election Commission Statement” .

Green, Donald P. and Jonathan S. Krasno. 1988a. “Salvation for the Spendthrift Incumbent: Reestimating the Effect of Campaign Spending in House Elections.” American Journal of Political Science 32:884-907.

________. 1990. Rebuttal to Jacobson’s “New Evidence for Old Arguments.” American Journal of Political Science 34:363-372.

Herrnson, Paul S. 2000. Congressional Elections, 3rd ed. Washington, DC:CQ Press.

Jacobson, Gary C. 1993. “The Misallocation of Resources in House Campaigns.” In Lawrence C. Dodd and Bruce I. Oppenheimer, eds., Congress Reconsidered, 5th ed., Washington, DC:CQ Press.

________. 1997. The Politics of Congressional Elections, 4th ed., New York:Longman.

Krasno, Jonathan S. 1991. Challengers, Competition, and Reelection. New Haven:Yale University Press.

Leyden, Kevin M. and Stephen A. Borrelli. 1990. “Party Contributions and Party Unity: Can Loyalty Be Bought?” Western Political Quarterly 43:343-365.

Poole, Keith T. 1998. "Recovering a Basic Space From a Set of Issue Scales." American Journal of Political Science, 42:954-993.

Rohde, David, W. 1979. “Risk Bearing and Progressive Ambition: The Case of Members of the United States House of Representatives.” American Journal of Political Science 23:1-26.

Schlesinger, Joseph. 1966. Ambition and Politics Chicago: Rand McNally.

Sorauf, Frank J. and Scott A. Wilson. 1994. “Political Parties and Campaign Finance: Adaptation and Accommodation Toward a Changing Role.” In L. Sandy Maisel, ed., The Parties Respond, 2nd ed., Boulder, CO:Westview Press.

Van Natta, Don, Jr. and John M. Broder. 2000. “Democrats Build Edge in Cash Battle to Control House.” New York Times July 12, 2000.

Wilcox, Clyde. 2000. They Did it Their Way: Campaign Finance Principles and Realities Clash in Wisconsin in 1998.” In Michael A. Bailey, Ronald A. Faucheux, Paul S. Herrnson, and Clyde Wilcox eds., Campaigns and Elections. Washington, DC:CQ Press.

Table 1:

Party Receipts and Expenditures on Senate Campaigns, 1995-96 and 1997-98

(Source: )

1995-1996

| |N |Receipts |Party Contributions |Party Coordinated Expenditures |

|Senate |316 |$285,151,978 |$163,1617 |$19,920,139 |

| | | | | |

|Democrats |96 |126,487,021 |671,098 |8,886,096 |

|Incumbents |7 |35,109,799 |165,933 |1,330,580 |

|Challengers |42 |40,516,067 |214,406 |2,638,799 |

|Open |47 |50,861,155 |290,759 |4,916,717 |

| | | | | |

|Republicans |114 |157,736,872 |960,519 |11,034,043 |

|Incumbents |14 |46,739,485 |269,548 |4,005,871 |

|Challengers |32 |38,001,323 |208,624 |2,674,545 |

|Open |68 |72,996,064 |482,347 |435,3627 |

1997-1998

| |N |Receipts |Party Contributions |Party Coordinated Expenditures |

|Senate |270 |$287,558,984 |$817,1357 |$18,684,013 |

| | | | | |

|Democrats |79 |134,102,990 |302,478 |9,349,948 |

|Incumbents |15 |66,941,216 |132,170 |4,955,907 |

|Challengers |51 |43,616,521 |73,164 |3,201,967 |

|Open |13 |23,545,253 |97,144 |1,192,074 |

| | | | | |

|Republicans |109 |153,028,494 |514,657 |9,334,065 |

|Incumbents |14 |68,577,760 |211,703 |4,110,684 |

|Challengers |82 |70,258,720 |244,763 |4,117,024 |

|Open |13 |14,192,014 |58,191 |1,106,357 |

Table 2.

Senate Election Finance Receipts and Expenditures, 1990-1998

(Source: Federal Election Commission)

Yearly Total Total Receipts

|Year |Democrats |Republicans |

|1990 |$89,621,120 |$96,837,616 |

| |(82) |(69) |

|1992 |145,146,368 |122,423,960 |

| |(123) |(126) |

|1994 |133,559,136 |184,060,528 |

| |(106) |(119) |

|1996 |126,183,296 |158,782,528 |

| |(95) |(114) |

|1998 |134,153,104 |153,191,568 |

| |(79) |(109) |

Number of Races in parentheses

Yearly Total, Major Party Contributions

|Year |Democrats |Republicans |

|1990 |$541,972 |$953,423 |

| |(82) |(69) |

|1992 |701,653 |779,814 |

| |(123) |(126) |

|1994 |650,600 |743,551 |

| |(106) |(119) |

|1996 |698,415 |962,494 |

| |(95) |(114) |

|1998 |302,478 |514,657 |

| |(79) |(109) |

Number of Races in parentheses

Yearly Total Party Coordinated Expenditures

|Year |Democrats |Republicans |

|1990 |$5,182,717 |$7,716,912 |

| |(82) |(69) |

|1992 |11,869,036 |16,516,738 |

| |(123) |(126) |

|1994 |12,966,360 |11,235,671 |

| |(106) |(119) |

|1996 |8,804,079 |11,053,476 |

| |(95) |(114) |

|1998 |9,595,222 |5,045,563 |

| |(79) |(109) |

Number of Candidates in parentheses

Table 3.

Tobit Estimates: Major Party Contributions and Coordinated Expenditures, 1990-98.

Dependent Variable: Ratio of Contributions and Coordinated Expenditures to Maximum Allowed

|Variable |I |II |III |IV |V |VI |

| | |Democrats |Republicans | |Democrats |Republicans |

|Constatnt |1.49 |0.25 |1.56 |1.29 |0.43 |1.40 |

| |(5.57) |(0.50) |(4.84) |(5.41) |(-0.87) |(4.90) |

|NOMINATE |-0.45 |-0.50 |-0.19 |-- |-- |-- |

| |(-0.95) |(-0.64) |(0.32) |-- |-- |-- |

|DISTANCE |-- |-- |-- |1.00 |-1.23 |1.79 |

| |-- |-- |-- |(1.49) |(-1.04) |(2.22) |

|THREAT |0.46 |0.60 |0.34 |0.47 |0.63 |0.35 |

| |(4.64) |(4.43) |(2.37) |(4.82) |(4.74) |(2.47) |

|INCUMBENT |-0.56 |-0.06 |-0.33 |-0.60 |0.01 |-0.79 |

| |(-2.60) |(-0.12) |(-2.64) |(-2.75) |(0.02) |(-3.06) |

|OPEN |-0.02 |0.63 |-0.33 |-0.12 |0.69 |-0.43 |

| |(-0.09) |(1.31) |(-1.02) |(-0.47) |(1.43) |(-1.40) |

|DEMOCRAT |-0.70 |-- |-- |-0.44 |-- |-- |

| |(-2.44) |-- |-- |(-3.41) |-- |-- |

|N |183 |88 |95 |183 |88 |95 |

t-statistic listed in parentheses

TABLE 4.

TOBIT ESTIMATES INDEPENDENT EXPENDTITURES, 1996-8 (1998 Dollars)

|Variable |Independent |Independent Expenditures |Independent Expenditures |Independent Expenditures |

| |Expenditures For |For |Against |Against |

|Constant |-584070.3 |-362109.6 |-549301 |-263218.9 |

| |(-2.51) |(-1.98) |(-1.53) |(-1.24) |

|NOMINATE |1113395 |-- |921773.6 |-- |

| |(2.13) |-- |(1.20) |-- |

|Distance |-- |1147425 |-- |212252.8 |

| |-- |(1.67) |-- |(0.20) |

|Threat |168996.2 |158949.2 |38657.29 |23611.58 |

| |(2.06) |(1.791) |(0.41) |(0.24) |

|Democrat |461044.1 |-214808.2 |676997.7 |104547.3 |

| |(1.37) |(-1.45) |(1.30) |(.68) |

|Open |42363.9 |109760.9 |-246203.6 |-194110.7 |

| |(0.28) |(0.67) |(-1.24) |(-1.01) |

|Incumbent |-178530.8 |-231196.3 |-460418.1 |-465112.1 |

| |(-1.34) |(-1.50) |(-2.34) |(-2.26) |

| | | | | |

|N |69 |69 |69 |69 |

t-statistic listed in parentheses

-----------------------

*Paper prepared for presentation at the Annual Meetings of the American Political Science Association, August 31to September 3, 2000, Washington, DC. I would like to thank Robert Biersack and Sharon Snyder at the FEC for data and documentation, and Keith Poole for providing NOMINATE data. ©American Political Science Association

[1] For an in-depth investigation of the Feingold-Neumann race, see Wilcox (2000).

[2] The terms ideology and policy preferences are used interchangeably.

[3] In their decision in the case FEC v. Colorado Republican Federal Campaign Committee, the US Court of Appeals for the Tenth Circuit found that such limits violated the First Amendment, essentially striking down such limits within the Tenth Circuit. The FEC warns that if the Supreme Court overturns the lower court’s ruling, the decision will be retroactive meaning that candidates who did not abide by such limits would be liable for violating the statute (FEC, 2000b).

[4] Independent expenditures were “created” in the Supreme Court decision FEC v. Colorado Republican Federal Campaign Committee, June 26, 1996. “The Court found that an expenditure made by the Colorado Federal Campaign Committee – specifically, a radio ad attacking a Democratic Congressman who had announced his candidacy for an open Senate seat – had not been “coordinated” with any candidate, therefy did not count against the coordinated party expenditure or contribution limits for the party’s eventual nominee for the same seat. The Court found “uncontroverted direct evidence that [the] advertising campaign was developed by the Colorado Party independently and not pursuant to any general or particular understanding with a candidate”” (FEC,1996, page 17 – FEC Campaign Guide for Political Party Committees, 1996).

[5] Both parties continue to raise large sums of money. In a departure from previous election cycles, the Democratic Congressional Campaign Committee has outraised their Republican counterparts by a significant margin. As for June 30, 2000, the DCCC had $37.4 million in the bank, over $15 million more than the Republicans (Van Natta and Broder, 2000). Both amounts represent significant increases over the already large advantages they held on March 31, 2000 ($15 million for Democrats and under $10 million for Republicans) (Clymer, 2000).

[6] Leyden and Borrelli (1990) argue that Democrats are more likely to reward loyalty while Republican party expenditures are more likely to generate higher levels of party unity. Cantor and Herrnson (1997) find that Democrats appeared to reward members for past loyalty during the 1992 election cycle. Those few exceptions involve the practice of Democratic party organizations to direct funds to seemingly safe incumbents instead of toward competitive challengers whose plights might have been helped with such funds (Jacobson, 1993; Shively and Wilson, 1994).

[7] Bianco (1999) studies party contributions to Senate candidates in 1992 and 1994 to determine whether the DSCC rewarded incumbent Senators for garnering funds for the DSCC in addition to their individual campaigns, or whether the DSCC distributed funds to candidates who needed them the most.

[8] See Krasno (1991) for a thorough analysis of the basis for House-Senate election differences.

[9] Other than Henry Clay, rarely do individual who served in the Senate seek seats in the House of Representatives. From 1947 to 1996, 266 (134 Republicans and 132 Democrats) House members voluntarily left to seek a position in the Senate.

[10] The most common type of measures used to capture ideology are roll call voting measures. A number of interest groups compile voting scorecards for the US House and Senate. Though less common, a number of these interest groups also compile voting scores for legislators at the state level, primarily from big states.

[11] The site is found at .

[12] The data files are called “cansum” followed by the year of the election in question. For example, the 1989-90 contribution data is found in a file named cansum90.zip.

[13] I used the inflation calculator at the Bureau of Labor Statistics web page to determine my deflator. See for details.

[14] Tobit is the preferred estimation procedure because the data are censored. That is, campaign contributions in this case cannot be negative. Use of tobit is standard in analyzing party strategy in campaign fund distribution (Cantor and Herrnson, 1997; Damore and Hansford, 1999; Bianco, 1999).

[15] One possible problem with this data stems from the fact that independent expenditures are relatively new, hence have not been widely used by the parties. Among the 69 candidates included in my dataset from the 1996 and 1998 election cycles, only 12 candidates received help from their party in the form of independent expenditures for, while 9 candidates had the opposing party spending on independent expenditures against them.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download