Contributions and contributors in the 2004 presidential election cycle.

Panagopoulos, Costas ; Bergan, Daniel

2004 marked a very close presidential race accompanied by unprecedented mobilization and high rates of interest, attention, and participation by the electorate (Bergan et al. 2006). The 2004 election marked a record turnout of over 60 percent of eligible voters, or 122.3 million voters, representing a dramatic increase from the 2000 election which attracted 105.4 million voters (or 54.2 percent of the voting-eligible population) to the polls. Voter turnout in 2004 was the highest of any election since 1968. Interest in the election and attention to the campaign also spiked in 2004. In 2004, 85% of respondents to the National Election Studies (NES) poll claimed to care a lot about the campaign compared with 78% in 2000. Similarly, in 2004, 52% of respondents claimed to be "very much" interested in the campaign, up from 29% in 2000. There was a corresponding increase in the level of attention to the presidential election; 65% of respondents claimed to pay "a great deal" or "quite a bit of attention" to the election in 2004 versus 47% in 2000 (NES 2004, 2000).

The 2004 elections were also accompanied by increased contributing behavior; presidential candidates more than doubled total campaign funding receipts from the 2000 electoral cycle. In this article, we will explore the population of donors and attempt to explain the reason for the increase in contributors. The results suggest that campaign contributing, like voting behavior, was affected by the competitiveness of the presidential race: two ideologically opposed candidates were facing one another in what many perceived as a close election. The level of interest in and the overall competitiveness of the race worked in conjunction with new institutional features, such as technological innovations and campaign finance regulations, to increase the number of contributions during the 2004 campaign.

Many different forms of participation are marked by unrepresentativeness of participants versus the population as a whole in either political attitudes or demographic characteristics (Verba, Schlozman, and Brady 1995). A typical complaint about campaign contributions is that they skew the political contest away from the typical voter by favoring wealthy contributors. While this may have been the case in the 2000 election, in the 2004 election there appears to be a rough balance in ideological contributors. While contributors are not ideologically similar to the average individual, there is a rough balance of the number of contributors on the left and right. The results suggest that with more widespread contributing during a campaign, contributors appear to be more evenly balanced ideologically; this claim has been made concerning other types of behavior (Rosenstone and Hansen 1993) and appears to hold for campaign contributing.

We will draw on two datasets to explore the 2004 contributors. The first data set involves aggregate contribution data. Because the contributions are categorized by size and by type of contribution (individual contributions versus PAC contributions or public funds), we can make some claims about the nature of contributions in the 2004 presidential election cycle. The large percentage of small contributions during the 2004 election suggests that it is greater participation by individuals interested in the campaign driving the increase in contributions, rather than say groups interested in some form of quid pro quo.

The second data set we use to explore donors, the NES survey, includes questions about a number of demographic and ideological attributes of donors versus noncontributors which allow us to determine how representative donors are of the population as a whole. In addition, the data allow us to attempt to explain why some individuals give and others do not.

The next section will use aggregate data to explore donors. The following two sections will use NES data; the first of these will explore the importance of the effect of institutional features such as the Internet and new reforms on contributing behavior. The second of these will explore the representativeness of donors. These two sections will be followed by a summary and some conclusions. Overall, we find that during the 2004 campaign, individual contributing was widespread due to interest in the campaign, and contributors, while different from the average individual in terms of demographic characteristics and ideology, were roughly balanced ideologically between the two major parties.

Political Contributions in the 2004 Presidential Campaign

Presidential contenders in the 2004 election cycle raised $674 million, more than doubling total receipts from the 2000 cycle. Individual contributors alone donated over $600 million to presidential candidates during the 2004 campaign, representing nearly a threefold increase from the $213 million that candidates raised cumulatively from individuals during the 2000 presidential election cycle (Malbin 2006; see Table 1).

Democratic candidates cumulatively raised $347 million from individual contributors in 2004, up from $63 million in 2000, and George W. Bush raised $257 million in 2004, sizably more than a total of $151 million raised by all Republican contenders in 2000. This growth is impressive especially considering it was achieved during the first presidential election cycle in which the Bipartisan Campaign Reform Act (BCRA) was in effect. Analysts had long debated the effect BCRA might have on presidential fundraising, and many believed its provisions--the ban on soft money in particular--could hinder candidates' ability to solicit funds. BCRA also increased the maximum individual contribution limit from $1,000 to $2,000, a change that also appears to have had significant implications in 2004.

The context in which presidential fundraising occurred in 2004 was influenced substantially by the fundraising dexterity President Bush had previously demonstrated. Bush's $96 million in total receipts during the 2000 race had shattered all previous records (Magleby 2002), creating high expectations for Republican performance in 2004 and prompting Democrats to reconsider and retune fundraising strategies.

One Bush decision the two leading Democratic contenders decided to emulate was to reject public financing during the primaries, freeing both candidates from the spending and other restrictions that may have compromised their ability to be competitive during the primary phase of the campaign. Early fundraising success enabled Howard Dean to announce on November 8, 2003, that he would opt out of the matching fund system. The Dean campaign had raised nearly $15 million between July 1 and September 30, 2003--more than three times as much as Kerry's next highest intake of $4 million over the same quarter (Malbin 2006). Still, five days following Dean's announcement, Kerry declared he would follow suit. No other Democrats could afford to reject public funds, but the three main actors in the primary phase of the campaign (Bush, Kerry, and Dean) had all opted out of the public presidential funding system.

Another key Bush strategy Democrats found useful was to rely more on high-dollar fundraisers. At the center of Bush's 2000 fundraising success were his Pioneers, individuals who had agreed to raise a minimum of $100,000 for Bush in 2000. A total of 241 pioneers alone raised $24.1 million for Bush in 2000. In 2004, the Bush campaign turned to networking again to raise large sums. The campaign added a new level, Ranger, who committed to raising at least $200,000, and 95 Mavericks, under-forty year olds who pledged to raise $50,000. The Campaign Finance Institute (CFI) estimates that by August 2004, this fundraising network had raised nearly $80 million (Malbin 2006). After Super Tuesday, when Kerry had effectively clinched the Democratic nomination, the Kerry campaign raised over $40 million from 266 Vice-Chairs (who raised $100,000 each) and 298 Co-Chairs ($50,000), a network modeled after Bush's (Public Citizen 2004).

Democrats also took another page from the Bush fundraising playbook by focusing on large donors. While both parties raised unprecedented sums from small (under $200) donors in 2004, the lion's share of funds for both parties was collected from large ($1,000+) contributors. Over the course of the campaign, 44 percent of Kerry's total receipts from individuals came from large donors. In fact, nearly one quarter of total Kerry receipts from individuals came from donors who contributed $2,000 or more. The source of 58 percent of Bush's total individual receipts was large donors, and 44 percent of Bush's individual contributions came from donors who contributed $2,000 or more. Large donors were especially important for the Bush campaign in early fundraising, accounting for 79 percent of the $129.5 million the Bush campaign had raised from individuals as of December 31, 2003. All in all, half of the total amount of individual contributions collected in 2004 came from large donors, and one third of total individual receipts were contributed by $2,000+ donors.

The nature of presidential fundraising in the 2004 election cycle was notably different in several key respects. One of the main stories to come out of the 2004 presidential fundraising cycle is that the race appears to have attracted many newcomers. Systematic analysis of Federal Election Commission (FEC) records conducted by the CFI reveals that only 30 percent of the 93,865 individuals who contributed more than $200 to Bush in 2000 donated to him again in 2004. Only 31 percent of the 61,116 donors who gave maximum contributions of $1,000 to Bush in 2000 gave any disclosed ($200+) amount to him in 2004. Michael Malbin, director of the CFI, observes that more than two thirds of Bush's 2000 donors did not give again in 2004 (Malbin 2006). On the Democratic side, only 25 percent of Al Gore's $200+ donors, and 21 percent of Bill Bradley's, gave more than $200 to any presidential candidate in 2004. Thus, the 2004 election cycle attracted many new donors. FEC records indicate 150,722 donors contributed sums greater than $200 to the Bush campaign in 2004. CFI analyses of these data reveal four out of every five (118,807 or 79 percent) were new donors or had not given in the 2000 campaign (Malbin 2006).

Of course, the big fundraising story of the 2004 presidential campaign was the unprecedented level of participation of small (under $200) donors. In 2000, small donors accounted for a total of $53.7 million in fundraising, or 25 percent of overall receipts from individuals. One fifth of Gore's fundraising ($6.8 million) and 16 percent ($14.8 million) of Bush receipts were in amounts less than $200. In 2004, over one third (34 percent) of total receipts from individuals came from small donors. Howard Dean's successful efforts to harness the power of the Internet to solicit contributions--especially in small amounts--may have helped to fuel this phenomenon. Dean raised over $30 million in small donor contributions. However, the other major candidates raised large quantities from small donors as well. Over the course of the campaign, Kerry raised $79.6 million (or 37 percent of his total individual receipts) from small donors, the bulk of which ($57 million) came in after Super Tuesday. Thirty-one percent of Bush's overall individual receipts ($78.7 million) came from under-$200 contributors (Malbin 2006).

Direct contributions to candidates are but one option available to donors. Political parties also engage in aggressive fundraising campaigns during presidential elections, and the national committees (Republican National Committee, RNC, and Democratic National Committee, DNC) were required to adapt to the new regulatory environment established by BCRA. After all, many of BCRA's main provisions, such as the ban on soft money, were designed to address shortcomings in party financing. Despite the fact that BCRA also increased limits for party contributions (from $20,000 to $25,000 per year to a national committee), many analysts believed BCRA would weaken the parties, or, at the very least, diminish their ability to help finance national elections. Moreover, experts felt BCRA would provide a partisan advantage to the Republican party, which had historically been more successful than the Democratic party at raising hard money contributions.

The new campaign finance rules prompted both political parties to modify their fundraising strategies. Emulating successful tactics pioneered by the presidential candidates, both national committees invested heavily in programs to attract new donors and to encourage previous donors to step up to the plate. They also pursued high-dollar contributions through a volunteer fundraising network. The DNC created the Patriots program (individuals who raised a minimum of $100,000 for the party during the 2004 cycle) and the Victory 2004 Trustees program (individuals who raised at least $250,000 for the party between May 1 and July 1, 2004). By the time the DNC hosted its nominating convention in 2004, the party had successfully recruited at least 17 Trustees and 188 Patriots to help with party fundraising (Corrado 2006). The RNC relied on similar donor programs, expanding its Republican Eagles and Team 100 programs, and establishing a Super-Rangers program for individuals who raised at least $300,000 for the party in addition to the $200,000 minimum they were required to raise for the Bush campaign. Over the duration of the campaign, the RNC had attracted 104 Super-Rangers (Corrado 2006).

In the end, the political parties did not collapse under BCRA. On the contrary, the national committees appear to have flourished, raising more money in hard dollars alone than they had raised in hard and soft money combined in any previous election cycle (Corrado 2006; see Table 2). Combined, national party committees raised over $1.2 billion in 2004, about $164 million more than they had raised in 2000 in both hard and soft money contributions (1) (Corrado 2006).

Despite expectations that BCRA would cripple the Democrats, the party succeeded in narrowing the fundraising gap substantially in 2004. Even as the Democrats trailed Republicans in total contributions overall by over $75 million ($576.2 million compared to $657.1 million, respectively), the gap was the smallest in more than twenty years (Corrado 2006). Moreover, the DNC outraised the RNC for the first time since the modern campaign finance era began in 1974 (Corrado 2006). The DNC's $394.4 million intake was more than $2 million higher than the RNC's total in 2004.

Part of the political parties' success in adapting to the new campaign finance rules in 2004 was their ability to identify, target, and mobilize new contributors and small donors. Anthony Corrado observes that "[t]he scale of the increase in donor participation was historic by any standard" (Corrado 2006). Corrado estimates that the DNC increased its direct mail donor base seven-fold in 2004 (from 400,000 in 2000 to 2.7 million). Four million donors made their contributions over the Internet. The DNC reports raising more than $248 million in small contributions in 2004, $35 million more than in 2000. Analysts estimated the RNC increased its small donor contributions by 75 percent in 2004 (Corrado 2006).

Despite attention to small donors, the national committees did not neglect large donors. FEC records indicate the RNC and DNC combined raised $104.2 million in maximum allowable contributions in 2004, up from a total of $23.7 million of hard money maximums in 2000. Corrado (2006) estimates that the RNC and DNC received at least 4,168 maximum contributions in 2004, compared to 1,185 maximum hard money donations in 2000, translating into $20.8 million in hard dollar receipts.

Many expected that BCRA's soft money prohibition would encourage former soft money donors--especially corporations (through PACs or employees identified with the organization)--to direct their contributions to 527 groups. While analysts observe some movement in that direction, the evidence suggests that the vast majority of former soft money donors did not increase their contributions to independent committees organized under section 527 of the tax code (Boatright et al. 2006). Among business donors (who comprise the bulk of soft donors), a few gave a lot more money, but most cut back. Large, publicly traded corporations, a major source of soft money in previous cycles, participated with lower frequency in 2004, although labor unions and a few "mega-donors" (individuals who contributed over $1 million to 527 groups in 2004) increased giving substantially (Boatright et al. 2006). In the end, 527 groups active in federal campaigns raised and spent $424 million in 2004, a sum that reflects substantial increases in both the overall number and average size of contributions by individual donors (Weismann and Hassan 2006).

For a variety of reasons, including the new campaign finance environment established by the BCRA and operating for the first time in a presidential campaign cycle in 2004 and a highly polarized and competitive presidential campaign, patterns of giving in the 2004 presidential election appear to differ from previous cycles and beg close scrutiny. Initial evidence suggests the nature of the donor pool may have differed in significant ways in 2004. The remainder of this article uses available survey data to evaluate the characteristics and motivation of 2004 presidential donors.

Reforms, Technology, and Contributions in the 2004 Campaign

Media coverage of campaign fundraising during the 2004 campaign focused on a number of legal and technological innovations that were said to affect the 2004 elections. The first unique feature that generated some attention during the campaign was the role of the BCRA which banned soft money, or unlimited contributions from the parties. Other developments in fundraising during the campaign stemmed in part from this reform; substantial media attention was devoted to 527s, groups outside of the reach of the campaign finance regulations that legally raise money without registering with the FEC. Another innovation during the campaign was the use of email and the Internet by the campaigns, most notably by Dean's campaign, to raise funds.

Survey data suggest that campaign contributing was fairly widespread in the 2004 election. Table 3 presents the percentage of respondents in the NES survey who claimed to contribute to a candidate or party, as well as to which side the contributions were made. The table also presents the percentage who claimed to give to a political interest group seeking to influence the election, such as America Coming Together or Emily's List. Table 4 shows the percentages reporting contributions during the 2000 election cycle.

The rightmost columns of Table 3 show that campaign contributing during the 2004 elections was fairly widespread; about 17 percent of respondents reported contributing to one or more of the parties, candidates, or political groups during the 2004 election cycle, up from 12 percent in 2000. (2) It appears that most of this increase is attributable to an increased number of contributions to Democratic candidates; this increase brought about rough parity in the number of individual contributors to the two major parties in the 2004 election cycle. This will be discussed in greater detail below in the ideology of donors.

The increase in the percentage of donors may be due, in part, to the activities of nonparty groups (such as Swift Boat Veterans for Truth), but the percentage of respondents in the NES data who claimed to give to a group other than a party increased by only about 2 percentage points. This increase is not large relative to the 3 percentage point increase in contributions to parties and the 4 percentage point increase in contributions to candidates. Much coverage discussed the role of these 527 groups in the aftermath of reforms that would restrict contributions to parties; however, the increase in the percentage given to parties was larger than that to all other groups of recipients.

Internet giving also grew in the 2004 campaign. One study estimates that the number of individuals contributing to a campaign grew from 2 million in 2000 to 4 million in 2004 (Pew Research Center for the People & the Press 2005).

The data are consistent with the thesis that contributions increased as the campaigns attempted to raise more funds from small donors in response to the ban on soft money; the percentage of the sample giving to groups, parties, and candidates increased between 2000 and 2004. This is consistent with the aggregate data discussed above, as Bush and Kerry raised a large percentage of funds from donors giving less than $200.

The survey data do not allow us to explore the average size of contributions accurately, as the surveys do not ask respondents about the size of their campaign contributions. However, the NES survey data, with detailed information on demographics and political attitudes, do allow a detailed exploration of the relationship between socioeconomic status, political attitudes, and the decision to make a campaign contribution. The next section will use the NES data to explore a long-standing concern in political fundraising: the representativeness of donors in the 2004 campaign.

In addition to technological innovations such as the use of the Internet to raise funds, the large number of contributions was also motivated by contextual features of the campaign (the closeness of the race and increased interest in the election) and attributes of the donors themselves. The next section will discuss the socioeconomic characteristics of donors as well as their attitudes toward the 2004 elections in addition to campaign activity by the candidates, political parties, and other groups in 2004. The fact that political attitudes and other characteristics of donors may have driven contributing behavior raises the possibility that donors have political beliefs unrepresentative of the population as a whole. This possibility is explored below.

Campaign Donors and Political Representation

A perennial fear about campaign contributions is that contributors are unrepresentative of the population as a whole. This section will explore the descriptive and ideological characteristics of donors. Campaign contributors in the 2004 campaign were descriptively unrepresentative of the population as a whole. Contributors were also closer to the political extremes than the average citizen, but during the 2004 election contributors were roughly balanced, with comparable proportions on both the left and right giving. Although this says little about the amounts contributed from either side of the political spectrum, in 2004, the number of contributors from the left and right were roughly balanced.

Previous research has established that campaign contributors are unrepresentative in descriptive characteristics: individuals who contribute to political campaigns are disproportionately white, male, and wealthy (Francia et al. 2003). Table 4 presents zero order correlations of different forms of participation with education and income. (3) The forms of participation include voting, influencing the vote of another person, attending party meetings, displaying a sign or bumper sticker, or doing other work for a campaign; correlations for both 2004 and 2000 are included. The results are the same for both elections cycles: income and education are most highly correlated with contributions; the only act of participation that is more highly correlated with education and income is voting. Other acts of participation, such as displaying a sign or bumper sticker, are not significantly correlated with education and income while others, such as attending meetings or doing other work for a campaign, are not as highly correlated with income and education.

Table 5 presents probit regressions of donors and includes some contextual features of campaign contributors. Model 1 regresses contributing behavior on socioeconomic characteristics and Model 2 shows that features of the 2004 elections were driving the number of campaign contributions. The models include age, age squared, respondent gender, race, and whether or not the individual considered themselves a strong partisan; these socioeconomic variables have been associated with political participation in general and campaign contributions in particular (Verba, Schlozman, and Brady 1995; Francia et al. 2003). The models also control for whether or not the individual resided in a battleground state (4) in the year in question and whether or not they report being contacted by a party or anyone else during the course of the campaign. The second model includes variables for interest in the campaign, including the respondent's perceived distance between the candidates, whether or not the respondent thinks the election "will be close," is "very interested" in the campaign, and cares "a great deal" about the election. Models 3 and 4 present the results for identical regressions for the 2000 election cycle. Aside from the expected influence of income on contributing behavior, the models show that general interest in the campaign spurred contributing behavior. In both election cycles, Models 2 and 4 show that interest in the campaign increased the probability of contributing. The proportion of those claiming that they were very interested increased from .29 to .52. The latter increase is particularly important in explaining the increase in donors, as interest in campaigns increases the probability of making a campaign contribution (Table 5, column 2). In 2004, the effect of caring "a great deal" about the election is statistically significant; this variable is not significant in 2000. Comparing Models 2 and 4 in Table 5, the distance between the presidential candidates exerts a marginally significant impact on contributing in 2004 and has the wrong sign in 2000. Paired with the fact that between 2000 and 2004 the average distance between presidential candidates doubled from 1.6 to 3.2, the perception of the distance between the two major presidential candidates may partially explain the increase in presidential fundraising among individual donors. The results suggest that greater excitement about the 2004 campaign resulted in a larger number of individual contributions.

In both 2000 and 2004, being contacted by parties and/or candidates increased the probability of contributing. This supports the theory that for many forms of participation, including contributing behavior, mobilization by candidates and groups is a key factor in political behavior (Rosenstone and Hansen 1993). The fact that party contacts increased between the 2000 and 2004 campaigns (Bergan et al. 2006) partially explains the increase in funds raised by the major party candidates.

What is clear from Tables 4 and 5 is that donors, like voters, are not representative of the population in terms of socioeconomic status. Income and education are related to campaign contributing behavior. This relationship persists while controlling for partisanship and interest in campaigns. This could affect policy if donors have significantly different policy views from the public as a whole. This section has shown, however, that in 2004, there was a larger percentage of Democratic donors that roughly balanced out the percentage giving to the Republican and Democratic parties and candidates (although not necessarily the amount contributed). We will now explore the policy preferences of donors during the 2004 campaign. The data show that contributors were unrepresentative of the typical citizen in terms of policy preferences, but donors to either side of the political spectrum appeared to be roughly balanced on either side of the spectrum.

The ideological unrepresentativeness of donors has been less well explored than their descriptive characteristics. There are two ways that political donors could distort the political process. The first is that political donors may, due to their demographic differences, have political beliefs that are biased away from the average individual's. Verba, Schlozman, and Brady (1995) and Rosenstone and Hansen (1993) hypothesize that disproportionate participation among those with high socioeconomic status will slant the policymaking process by biasing political communications to public officials. Griffin and Newman (2005) show that voters have policy preferences that differ systematically from nonvoters and do in fact influence policy more than nonvoters. Bartels (2002) has shown that legislators are more responsive to constituents with higher income levels.

A second way that campaign contributors may distort the political process is by pulling the two major parties from the ideological center. Simple economic theories of two-party competition predict convergence at the ideal point of the median voter (Downs 1957); models that include individual campaign donors have suggested that activists of the two major parties, including campaign contributors, pull the candidates from the ideological center in search of resources necessary to win election, such as campaign funds (Aldrich 1983; Aldrich with McGinnis 1989; Moon 2004).

Figure 1 presents a histogram of the ideological positions of noncontributors and contributors on a seven-point ideological scale. The moderates in 2004 (individuals in the middle three ideological categories) are underrepresented. This underrepresentation may be exaggerated by the fact that many noncontributors may place themselves at the ideological center because they do not know where they fit on the scale. However, even considering this possibility, the underrepresentation of moderates is striking; the percentage of true moderates (a score of 4 on the 7-point scale) among contributors is half the percentage of moderates among donors. The extremes of the ideological distribution are overrepresented among donors; there are higher percentages of self-proclaimed liberals, conservatives, and extreme liberals and conservatives among donors. The figure suggests that the ideological unrepresentativeness is not unidirectional; both extremes of the ideological distribution are overrepresented.

[FIGURE 1 OMITTED]

Figure 2 shows that this was not the case in the 2000 election cycle. The mean ideological rating for contributors is 4.4 versus 4.2 for noncontributors, a difference significant at p < .10. The mean ideological rating for contributors to candidates is 5.2 versus 4.2 for noncontributors (where 7 represents extreme conservatives and 1 represents extreme liberals); this difference is statistically significant at alpha = .001 (two-tailed). In 2004, with a substantially greater percentage of individuals giving to Democratic candidates, the ideological balance of contributors moved to the left. In 2004, the mean ideology of contributors was 4.3 for both contributors and noncontributors.

[FIGURE 2 OMITTED]

Table 6 displays the ideological positions of donors on a number of specific issues. Donors to each of the major parties consistently have preferences that are statistically distinguishable from noncontributors. These patterns suggest that the second kind of participatory bias, in which activists from either party are more active than the center, is a more accurate description of patterns of contributions than a unidirectional bias.

What about the policy priorities of contributors? Do donors place weight on the same issues as noncontributors? The NES survey asks respondents to name the most important problem facing the nation. All categories with mentions by more than 10 percent of respondents were assigned an indicator variable equaling one if the respondent mentioned the issue as the most important and zero otherwise. Table 7 displays a probit regression of contributions on the same demographic and political variables used in Table 5 in addition to the indicator variables for most important problem.

The regression shows that contributors do have different issue priorities than noncontributors. The results suggest that contributors were more in tune with the issues highlighted by major party candidates during the campaign; the coefficients for moral issues, the economy, and trust in leaders are statistically significant. All of these were issues stressed by the candidates and covered in the media during the campaign. Surprisingly, however, the coefficients for Iraq and terrorism were not significant.

How does the representativeness of participants in campaign contributing compare to the representativeness of participants in other types of political behavior? Table 8 displays seven point ideological self-placements for participants and nonparticipants in different types of political behavior. In 2004, as discussed above, the ideology of contributors and noncontributors was the same; in 2000, contributors are more conservative that noncontributors by a small amount (significant at p < .01). Voters tend to be more conservative than nonvoters. Participants and nonparticipants of other types of behavior listed have roughly the same ideological self-placement, with the exception of doing "other work" for a campaign in 2004, where participants were more liberal than nonparticipants. The data suggest that with wider participation (as in 2004) contributors are roughly balanced ideologically. This does not seem to be the case for voting, however.

Conclusion

The number of individual contributors in the 2004 election greatly increased relative to 2000. These individual contributions came from a large percentage of new donors and small contributors. Despite media attention to the role of the Internet, 527s, and the BCRA, the number of individual contributors seems to be influenced as much by overall interest in the election than technology or reforms. This suggests that campaign contributions are similar to other forms of participation.

With the increase in campaign contributions in 2004 came a rough parity between the two major parties. As with voters, donors are wealthier and more educated than the average individual. However, this does not necessarily result in an ideological bias among contributors. In fact, with widespread contributing in the 2004 election, there appeared to be a sort of ideological balance on either side of moderates; this distinguishes the election from the 2000 electoral cycle.

In addition, as a large percentage of donors were first-time donors and gave small amounts, it appears that there is a large amount of money given that does not involve any sort of quid pro quo between candidates and donors. In 2004, there was a lot of money in the political system motivated by interest in the campaign rather than narrow economic benefits.

References

Aldrich, John H. 1983. A Downsian spatial model with party activism. American Political Science Review 77(4): 974-90.

Aldrich, John H., with Michael D. McGinnis. 1989. A model of party constraints on optimal candidate positions. Mathematical and Computer Modeling 12(4-5): 437-50.

Bartels, Larry M. 2002. Economic inequality and political representation. Working paper. Retrieved January 21, 2006 from http://www.princeton.edu/-bartels/economic.pdf.

Bergan, Daniel E., Alan Gerber, Alan Green, and Costas Panagopoulos. 2006. Grassroots mobilization in the 2004 campaign. Public Opinion Quarterly 69(5): 760-77.

Boatright, Robert G., Michael Malbin, Mark J. Rozell, and Clyde Wilcox. 2006. Interest groups and advocacy organizations. In The election after reform: Money, politics and the Bipartisan Campaign Reform Act, edited by Michael Malbin. Lanham, MD: Rowman and Littlefield.

Corrado, Anthony. 2006. Party finance in the wake of BCRA: An overview. In The election after reform: Money, politics and the Bipartisan Campaign Reform Act, edited by Michael Malbin. Lanham, MD: Rowman and Littlefield.

Downs, Anthony. 1957. An economic theory of democracy. New York: Harper.

Francia, Peter, John C. Green, Paul S. Herrnson, Lynda W. Powell, and Clyde Wilcox. 2003. The financiers of congressional elections: Investors, ideologues and intimates. New York: Columbia University Press.

Griffin, John D., and Brian Newman. 2005. Are voters better represented? Journal of Politics 67(4): 1206-27.

Magleby, David B., ed. 2002. Financing the 2000 election. Washington, DC: Brookings Institution Press.

Malbin, Michael. 2006. A public funding system in jeopardy: Lessons from the presidential nominating contest of 2004. In The election after reform: Money, politics and the Bipartisan Campaign Reform Act, edited by Michael Malbin. Lanham, MD: Rowman and Littlefield.

Moon, Woojin. 2004. Party activists, campaign resources and candidate position taking: Theory, tests and applications. British Journal of Political Science 34(4): 611-33.

Pew Research Center for People & the Press. 2005. The Internet and the Campaign 2004. Retrieved on March 9, 2006 from http://www.pewinternet.org/.

Public Citizen. 2004. The importance of bundlers to the Bush and Kerry campaigns: Post-election summary of findings. Washington, DC: Author.

Rosenstone, Steven J., and John Mark Hansen. 1993. Mobilization, participation, and democracy in America. New York: Macmillan.

Verba, Sidney, Kay Lehman Schlozman, and Henry E. Brady. 1995. Voice and equality: Civic voluntarism in American politics. Cambridge, MA: Harvard University Press.

Weissman, Steve, and Ruth Hassan. 2006. BCRA and the 527 Groups. In The election after reform: Money, politics and the Bipartisan Campaign Reform Act, edited by Michael Malbin. Lanham, MD: Rowman and Littlefield.

COSTAS PANAGOPOULOS

DANIEL BERGAN

Yale University

(1.) This includes $53 million in leftover funds from the presidential primary phase that was transferred to the national committees.

(2.) Note that these figures may include contributions to state and local candidates, party organizations, or other groups.

(3.) The NES data include a seven-category measure of education. The respondent income variable was recoded into a (our-category variable to allow comparability across electoral cycles; the categories are: 0 to $24,999, $25,000 to $49,000, $50,000 to $104,999, and $105,000 and up.

(4.) Battleground states were obtained from http://www.CNN.com; in 2004 these states included CO, FL, IA, ME, MI, MN, MO, NV, NH, NM, OH, OR, PA, WV, and WI, and in 2000 included AZ, AR, DE, FL, IL, IA, LA, ME, MI, MO, NV, NH, NM, OH, OR, PA, TN, WA, and WI.

Costas Panagopoulos and Daniel Bergan are postdoctoral fellows at the Institution for Social and Polky Studies at Yale University.

TABLE 1

Presidential Fundraising (Individual Contributions in Millions
of Dollars) 2000 and 2004

 Total Total Total $1,000+
 Net Public Individual Contrib.
Candidate Receipts Dollars Contrib.

2004
Wesley Clark 28.4 7.6 17.3 8.5
Howard Dean 51.1 0 51.1 9.7
John Edwards 31.7 6.6 21.6 15.4
Richard Gephardt 21.2 4.1 14.3 10.2
Bob Graham 5 0 4.4 3.2
John Kerry 234.6 0 215.5 94.2
Dennis Kucinich 12.4 3 7.9 .9
Joe Lieberman 18.5 4.3 14 10.5
Carol Mosley Braun 0.6 0 0.5 .3
Al Sharpton 0.7 0 0.5 .3
Democrats 404.2 25.6 347.2 153.3
George W Bush 269.6 0 257.4 148.7
Republicans 269.6 0 257.4 148.7
Total 673.8 25.6 604.6 301.9

2000
Bill Bradley 49 12.5 29.2 18.8
Al Gore 49.4 15.5 33.9 21.3
Democrats 98.4 28 63.1 40.2
Gary Bauer 16.8 4.9 7.6 1.1
George W. Bush 95.6 0 92.3 61.7
Elizabeth Dole 5.1 0 5 3.1
Steve Forbes 47.9 0 5.5 1.7
Alan Keyes 4.8 4.2 8 .4
John McCain 56.3 14.5 28.1 9.7
Dan Quayle 7.8 2.1 4.1 1.7
Republicans 234.3 25.7 150.7 79.4
Total 332.7 53.7 213.8 119.6

 $1,000+ $2,000+
 as % as % $200-
 Total $2,000+ Total $999
Candidate Indiv. Contrib. Indiv. Contrib.

2004
Wesley Clark 49 5.3 31 3.4
Howard Dean 19 3.9 8 11.1
John Edwards 71 10.4 48 3.5
Richard Gephardt 72 6.4 45 2.3
Bob Graham 73 2.2 50 .9
John Kerry 44 52.4 24 43.1
Dennis Kucinich 11 0.4 5 1.5
Joe Lieberman 75 6.2 44 2.5
Carol Mosley Braun 51 0.1 27 .1
Al Sharpton 64 0.2 45 .1
Democrats 44 87.7 25 68.6
George W Bush 58 113.3 44 31.6
Republicans 58 113.3 44 31.6
Total 50 201 33 100.2

2000
Bill Bradley 65 N/A N/A 6.9
Al Gore 63 N/A N/A 6.3
Democrats 64 N/A N/A 13.3
Gary Bauer 15 N/A N/A 2
George W Bush 67 N/A N/A 17.5
Elizabeth Dole 61 N/A N/A 1.1
Steve Forbes 31 N/A N/A 1
Alan Keyes 5 N/A N/A 1.2
John McCain 34 N/A N/A 7.5
Dan Quayle 41 N/A N/A .7
Republicans 53 N/A N/A 31
Total 55 N/A N/A 44.2

 $200- Less Less Than
 $999 as Than $200 as
 % Total $200 % Total
Candidate Indiv. Contrib. Indiv.

2004
Wesley Clark 20 5.4 31
Howard Dean 22 30.6 60
John Edwards 16 3 14
Richard Gephardt 16 1.8 13
Bob Graham 19 .4 9
John Kerry 20 79.6 37
Dennis Kucinich 19 5.5 70
Joe Lieberman 17 1.2 9
Carol Mosley Braun 27 .1 24
Al Sharpton 24 .1 13
Democrats 20 127.7 37
George W Bush 12 78.7 31
Republicans 12 78.7 31
Total 17 206.4 34

2000
Bill Bradley 23 3.8 13
Al Gore 18 6.8 20
Democrats 21 10.6 17
Gary Bauer 26 4.6 60
George W Bush 19 14.8 16
Elizabeth Dole 21 .9 19
Steve Forbes 17 3 53
Alan Keyes 16 6.4 80
John McCain 26 11.5 41
Dan Quayle 16 1.8 45
Republicans 20 43.1 29
Total 20 53.7 25

N/A, not applicable.

Sources: Federal Election Commission; Malbin (2006).

TABLE 2
National Party Committee Fundraising ($ Millions)

 2000 2004

Committee Hard Soft Total Hard/Total

DNC 124.0 136.6 260.7 394.4
DSCC 40.5 63.7 104.2 88.7
DCCC 48.4 56.7 105.1 93.2
Democrats 212.9 245.2 458.1 576.2
RNC 212.8 166.2 379.0 392.4
NRSC 51.5 44.7 96.1 79.0
NRCC 97.3 47.3 144.6 185.7
Republicans 361.6 249.9 611.5 657.1
Total 574.5 495.1 1069.6 1233.2

Note: DNC, Democratic National Committee; DSCC, Democratic
Senatorial Campaign Committee; DCCC, Democratic Congressional
Campaign Committee; RNC, Republican National Committee; NRSC,
National Republican Senatorial Committee; NRCC, National
Republican Congressional Committee.

Source: Federal Election Commission; Corrado (2006).

TABLE 3

Percent Contributing to Candidates, Parties, and Other Groups,
2000 and 2004

 Candidate Party Group Other Than Party Any Contribution

2004
Republican 5 5 -- --
Democrat 5 4 -- --
Total 10 9 6 17

2000
Republican 4 4 -- --
Democrat 2 3 -- --
Total 6 6 4 12

Note: Cell entries indicate percentages. Partisan leanings are not
available for groups other than parties.

TABLE 4
Correlations between Income, Education, and Various Forms
of Participation, 2000 and 2004

 Influence
 Other's
 Contribute Vote Vote

2004
Income .20 *** .18 *** .13 ***
N 974 974 974
Education .23 *** .28 *** .13 ***
N 1066 1066 1066

2000
Income .22 *** .19 *** .13 ***
N 1384 1385 1385
Education .18 *** .30 *** .13 ***
N 1550 1550 1551

 Display Other
 Attend Sign/Bumper Work for
 Meetings Sticker Campaign

2004
Income -.06 -.02 .00
N 816 974 974
Education .08 * .02 .08 **
N 896 1066 1066

2000
Income .10 *** .13 *** .03
N 1385 1385 1385
Education .10 *** .13 *** .06
N 1551 1551 1551

Note: Cell entries indicate Pearson's R correlations between
income or education and types of participation listed.

* Significant at .05; ** significant at .01; *** significant at .001.

TABLE 5
Probit Regressions of Contributions on
Respondent Characteristics

 2004

 Model| 1

Income .177 (.066) **
Education .150 (.037) ***
Age .016 (.019)
Age squared .000 (.000)
Female .038 (.114)
Black -.291 (.179)
Battleground state .152 (.118)
Contacted by party or "someone .437 (.114) ***
 else"
Strong party identification .627 (.107) ***
Distance between pres. candidates
Pres. election "will be close"
"Very much interested" in
 campaign
"Care a great deal" about election
Constant -3.18 (.485) ***
N 963
LR Chi square 151.00
Prob. > Chi square .000
Pseudo R square .171

 2004

 Model 2

Income .134 (.070) *
Education .123 (.040) **
Age .017 (.021)
Age squared .000 (.000)
Female .077 (.125)
Black -.389 (.213)
Battleground state -.116 (.127)
Contacted by party or "someone .335 (.122) **
 else"
Strong party identification .412 (.122) **
Distance between pres. candidates .073 (.042)
Pres. election "will be close" -.177 (.146)
"Very much interested" in .257 (.128) *
 campaign
"Care a great deal" about election .720 (.340) *
Constant -3.84 (.636) ***
N 818
LR Chi square 141.87
Prob. > Chi square .000
Pseudo R square .183

 2000

 Model 3
Income .186 (.091) *
Education .163 (.049) **
Age .035 (.027)
Age squared .000 (.000)
Female -.289 (.144) *
Black -.125 (.245)
Battleground state -.058 (.141)
Contacted by party or "someone .786 (.143) ***
 else"
Strong party identification .212 (.143)
Distance between pres. candidates
Pres. election "will be close"
"Very much interested" in
 campaign
"Care a great deal" about election
Constant -3.80 (.716) ***
N 785
LR Chi square 106.18
Prob. > Chi square .000
Pseudo R square .200

 2000

 Model 4

Income .219 (.098) *
Education .103 (.055)
Age .028 (.030)
Age squared .000 (.000)
Female -.299 (.155)
Black -.083 (.257)
Battleground state -.062 (.150)
Contacted by party or "someone .790 (.158) ***
 else"
Strong party identification .045 (.159)
Distance between pres. candidates -.006 (.062)
Pres. election "will be close" .091 (.062)
"Very much interested" in .562 (.155) ***
 campaign
"Care a great deal" about election .496 (.278)
Constant -3.97 (.846) ***
N 666
LR Chi square 109.59
Prob. > Chi square .000
Pseudo R square .226

Note: Standard errors are in parentheses.

* Significant at .05; ** significant at
.01; *** significant at .001.

TABLE 6

Political Attitudes of Contributors and Noncontributors, 2004

 Did Not Contribute Contribute
 Contribute to Democratic to Republican
 to Party or Party or Party or
 Candidate Candidate Candidate

Favor school vouchers .32 .23 .40
Big government too involved .39 .27 ** .70 ***
Free market .33 .25 * .54 ***
The less government the
 better .41 .26 ** .80 ***
Let gays adopt .48 .81 ** .31 ***
Let gays marry .35 .63 *** .11 ***
Favor laws to protect gays .76 .87 ** .58 ***
Government equality .74 .92 *** .33 ***
Favor affirmative action .18 .38 *** .04 ***
Increase aid to poor .59 .60 .24 ***
Increase spending: war on
 terror .42 .35 .64 ***
Favor government funds for
 abortion .37 .66 .23 **
Favor partial birth
 abortion ban .66 .29 *** .69
Favor investing Social .64 .25 *** .77 **
 Security

Note: Cell entries are proportions of respondents in category who
agree with policy.

* Indicates contributors to listed party are significantly different
from noncontributors at .05; ** significant at .01; *** significant
at .001.

TABLE 7

Probit Regression of Contributions on Individual
Characteristics and Respondent's "Most Important Problem"

Income .190 (.068) **
Education .148 (.038) ***
Age .009 (.020)
Age squared .000 (.000)
Female .066 (.117)
Black -.242 (.182)
Battleground state -.145 (.120)
Contacted by party or "someone else" .434 (.116) ***
Strong party identification .614 (.109) ***
Unemployment .075 (.368)
Health care -.057 (.488)
Terrorism .338 (.199)
Moral issues .844 (.362) *
Government spending/deficit .489 (.392)
Economy .497 (.244) *
Foreign policy .273 (.365)
Iraq .234 (.224)
Peace .451 (.351)
Defense .342 (.378)
Trust in leaders 1.01 (.468) *
Constant -3.40 (.511) ***
N 963
Likelihood Ratio chi square 163.05
Prob. > Chi square .000
Pseudo R square .185

Note: Standard errors are in parentheses.

* Significant at .05; ** significant at .01; *** significant at .001.

TABLE 8
Ideology of Nonparticipants and Participants

 Ideology: Ideology: N
 Nonparticipants Participants

2004
Voting 4.0 4.4 ** 824
Influence other votes 4.3 4.2 824
Other work for a campaign 4.3 3.9 * 824
Attend meeting 4.3 4.2 696
Display sign or sticker 4.3 4.2 824
Contribute 4.3 4.3 824

2000
Voting 4.0 4.3 *** 1238
Influence other votes 4.2 4.2 1238
Other work for a campaign 4.2 4.4 1238
Attend meeting 4.3 4.1 1238
Display sign or sticker 4.2 4.5 * 1238
Contribute 4.2 4.4 * 1237

Note: Cell entries indicate conservatism on a 7-point scale.

* Indicates significant difference (two-tailed) between the two
columns at .10; ** indicates a significant different at .05;
*** indicates a significant difference at .01.