[Congressional Record Volume 148, Number 23 (Wednesday, March 6, 2002)]
[Extensions of Remarks]
[Pages E261-E273]
From the Congressional Record Online through the Government Publishing Office [www.gpo.gov]

[[Page E261]]



                 BIPARTISAN CAMPAIGN REFORM ACT OF 2001

                                 ______
                                 

                               speech of

                           HON. ROBERT W. NEY

                                of ohio

                    in the house of representatives

                      Wednesday, February 13, 2002

       The House in Committee of the Whole House on the State of 
     the Union had under consideration the bill (H.R. 2356) to 
     amend the Federal Election Campaign Act of 1971 to provide 
     bipartisan campaign reform:

  Mr. NEY. Mr. Chairman, The following information I wish to submit for 
the Record on this matter of Campaign Finance Reform.

                Ney-Wynn Is The Effective Soft Money Ban

       Perhaps the best way to explain the difference between the 
     ``soft money bans'' is to elaborate on comments by the 
     President of Common Cause, Scott Harshbarger, panning Ney-
     Wynn. As you may know, Common Cause strongly supports Shays-
     Meehan, and was apparently heavily involved in drafting much 
     of its language. He incorrectly asserted that Ney-Wynn is 
     just like Senator Hagel's bill, and that Ney-Wynn continues 
     to let in unregulated and unlimited soft money. Both are 
     false--Ney-Wynn does ban (as opposed to cap) soft money for 
     Federal election activity.
       First, with respect to Senator Hagel's approach, I have 
     reviewed both bills, and am of the view that any comparison 
     to Ney-Wynn is an oversimplification. Senator Hagel's bill 
     merely put a limit on the amount of non-federal funds the 
     party committees could accept. It put no restriction 
     whatsoever on how the money could be spent, and would not 
     have dramatically altered how party committees currently 
     operate. Party committees would have still been free to run 
     so-called ``soft money issue ads'' and engage in other 
     similar activities. It would not have forced the party 
     committees to use federal money for federal election 
     activity.
       On the other hand, the Ney-Wynn bill would radically alter 
     how party committees on both sides of the aisle operate. 
     Unlike Hagel, it bans soft money for federal election 
     activity. It bans us from doing so-called ``soft money issue 
     ads.'' In short, Ney-Wynn actually accomplishes what the 
     reformers want--an end to party committee soft money being 
     used in Federal elections via the back door.
       With respect to soft money, Ney-Wynn bans the party 
     committees from using it for any Federal election activity. 
     As for the limited amount that party committees will be 
     allowed to accept, all the Ney-Wynn bill does is treat party 
     committees the same as corporate and union PACs, allowing us 
     to use limited soft money contributions for fundraising and 
     administration. And that's all--it can't be transferred, used 
     for issue ads or the like. In fact, Senator McCain himself 
     voted to let this use of soft money continue.
       Moreover, Ney-Wynn is consistent with the recent decision 
     in Colorado Republican II, where the Court said that ``[a 
     party committee] is in the same position as some individuals 
     and PACs.'' It also avoids the issue presented in Jacobus v. 
     Alaska and a recent opinion letter from the Attorney General 
     of New Hampshire (finding total contribution bans are 
     unconstitutional). So as to ensure that the Ney-Wynn ban is 
     truly a ban, even voter registration and get out the vote 
     conducted within 120 days of a federal election must be paid 
     for entirely with federal hard dollars. But critically, this 
     limitation still allows the party committees the ability to 
     do voter registration and other outreach.
       Several state laws already ban soft money in this way, most 
     notably Texas and New York, and to a certain extent, Florida 
     and California. States that have taken the Shays-Meehan 
     approach tend to be left-of-center, and eventually try some 
     form of public financing. The current Shays-Meehan actually 
     mandates that public financing be studied.
       As someone who has intimate knowledge of the financial and 
     political operations of party committees, I believe Ney-Wynn 
     is a radical change. It will force the parties to use hard 
     dollars for our activities, but still allow us the resources 
     to assist our candidates in getting their messages heard. The 
     party committees will remain able to drown out special 
     interest ads, albeit with hard dollars. In sum, Ney-Wynn 
     achieves what ought to be everyone's common goal: allowing 
     the candidate's voice to be the central voice in American 
     politics.
       Shays-Meehan will have the opposite effect--it simply 
     attempts to emasculate the political parties, and leave 
     candidates to fend for themselves. It does not make any 
     effort to ensure that parties continue to register voters and 
     involve people in the process. Once the Shays-Meehan 
     experiment inevitably makes matters worse, reformers will 
     then insist that public financing of all Federal elections is 
     the only option left.
       As for soft money, Shays-Meehan does not constitute a soft 
     money ban with respect to federal election activity. Contrary 
     to what has been repeated time and time again, ``soft money'' 
     is neither unlimited nor unregulated. Over 30 states have 
     passed limits or outright bans on corporate and union money, 
     including contributions to state political parties--laws that 
     Shays-Meehan actually preempt. In fact, when the full impact 
     of the two bills is analyzed, particularly in light of the 
     application (or preemption) of state law and the $30 million 
     soft money loophole, Ney-Wynn constitutes the more effective 
     soft money ban. It has the added advantage of requiring 
     disclosure of third-party issue ads that is consistent with 
     judicial precedent.
       Please contact me with any questions or concerns.

                                  ____
                                  

  [Institute of Governmental Studies and Citizens Research Foundation 
                      Policy Brief, July 6, 2001]

Assessing the Impact of a Ban on Soft Money: Party Soft Money Spending 
                         in the 2000 Elections

              (By Ray La Raja and Elizabeth Jarvis-Shean)

       This policy brief examines how national, state and local 
     parties of the Republicans and Democrats spent soft money in 
     the 2000 Elections. Our findings demonstrate that the state 
     parties, which receive about 83% of their soft money from 
     national party transfers, are the primary venue for soft 
     money spending. About 44% of state party soft money spending 
     went toward media activities, while 15% was invested in 
     mobilization and grassroots activities. Parties target their 
     media and mobilization spending in competitive states. The 
     Democrats rely more on soft money for campaign activity than 
     Republicans. Spending on all campaign activities--media, 
     mobilization and grassroots--has been increasing over the 
     past several election cycles. If soft money is banned--or 
     simply curtailed within 120 days of a general election--it is 
     likely that both media and party building activity will be 
     reduced significantly unless the parties can make up for the 
     shorfall with hard money.
       The purpose of this report is to furnish basic data about 
     soft money spending in the 2000 elections as a way to 
     understand the potential impact of campaign finance reform 
     legislation being debated in the 107th Congress. In 
     particular, we consider the effect of a ban on soft money, a 
     provision that remains the centerpiece of a bill sponsored by 
     Senators McCain and Feingold, and passed by the Senate on 
     April 2, 2001. The House of Representatives will soon 
     consider a similar version of the bill. Much of the debate 
     over reform considers the effect of eliminating soft money on 
     party activities. Will the parties be weakened? To what 
     degree are parties using soft money for issue ads? In this 
     report we assess how parties spent their soft money in past 
     elections as a way to understand the likely consequences of 
     banning or restricting soft money.
       Soft money includes funds that parties raise that lack the 
     contribution limits set by the Federal Election Campaign Act 
     of 1971 and its amendments. Under federal law, soft money may 
     be used for party building but not direct candidate support. 
     Advocates for banning soft money argue that its elimination 
     is essential for preserving the integrity of the electoral 
     system. Their underlying premise is that soft money corrupts 
     the political process by allowing wealthy donors to trade 
     political money for favorable treatment in policymaking in 
     Congress and the Executive branch. Some argue that even if 
     candidates are not corrupted, voters perceive that the 
     exchange is corrupt or that parties abuse campaign finance 
     laws by using funds illegally to help their candidates. Such 
     perceptions alienate voters from the political process and 
     undermine the legitimacy of the nation's political 
     institutions.
       Others argue, in contrast, that a ban on soft money will 
     damage American democracy. Citing several court decisions, 
     they claim that constraints on political activity run counter 
     to the 1st Amendment. Another line of argument contends that 
     eliminating soft money will weaken an essential political 
     institution in American democracy--the political parties. 
     Removing this resource will weaken parties relative to other 
     political actors such as interest groups, and reduce the 
     party's efforts to get voters to the polls. Rather than 
     reinvigorate political participation, the McCain-Feingold 
     reforms might actually reduce citizen activity.
       The arguments on either side deserve rigorous empirical 
     scrutiny. It appears, however, that Congress is poised to 
     enact legislation without considering some basic information 
     about soft money. Drawing on financial data about parties 
     released by the Federal Election Commission, we try to shed

[[Page E262]]

     some light on the uses of soft money. We are hardly prepared 
     to provide an in-depth analysis to address the claims of 
     either side in this upcoming reform debate in the House. 
     Instead, our goal is to provide an empirical foundation to 
     help policymakers consider carefully the ramifications of 
     their decisions.
       The questions we ask are simple, but to our knowledge they 
     have not been addressed adequately. How did parties spend 
     soft money in the 2000 elections? To what extent did they use 
     soft money to finance ``issue ads?'' How much soft money went 
     toward traditional voter mobilization efforts and other party 
     building activity? Did parties spend differently from prior 
     elections?
       Using data providing by the Federal Election Commission we 
     explore these questions about soft money spending. We 
     categorized more than 500,000 entries of
     National parties
       May raise or spend only hard money (i.e., limited 
     contributions, no labor or corporate contributions).
       May not make contributions to non-profits.
     State parties
       Must use hard money to fund any ``federal election 
     activities'' (defined as Get-Out-the-Vote, or voter 
     registration in the 120 days preceding an election) during a 
     federal election year.
       May fund ``federal election activities'' with soft money 
     capped at $10,000 from the same source if state laws permit.
     Candidates
       Banned from raising soft money for ``federal election 
     activities.''
     Non-profits
       National parties banned from making or soliciting 
     contributions to nonprofits; candidates banned from raising 
     soft money for non-profits for ``federal election 
     activities.''


                               Findings I

        How did parties spend soft money in the 2000 Elections?

       Parties at the federal, state and local level spent almost 
     half a billion dollars in soft money in the 2000 elections. 
     These funds were spent primarily by the state parties because 
     federal and state regulations are more permissive of soft 
     money spending at this level. The 100 major state parties--
     Democratic and Republican--spent approximately $340 million 
     in soft money. The national parties, in contrast, spent only 
     $136 million. The national parties, however, raised a good 
     portion
       Local parties (158 of them) spent only $4 million. It 
     should be noted that federal, state and local parties spend 
     additional soft money in non-federal elections. But because 
     these funds are not related in any way to a federal election 
     they do not have to be reported to the Federal Election 
     Commission. Therefore, the soft money data we collected 
     pertains only to campaign spending related to federal 
     elections. We should also point out that we are reporting 
     only soft money figures here. By law, parties are required to 
     match soft money with hard money for each activity, using 
     complex accounting guidelines provided by the Federal 
     Election Commission. If we included the hard money figures in 
     several of the subsequent tables, the spending in these 
     categories would be 40% to 50% higher.
       For national parties, most soft money (about 43%) is 
     invested in overhead and basic administrative costs of 
     maintaining the party headquarters in Washington. 
     Unsurprisingly, the next largest expenditure is for 
     fundralsing (approximately 39%), It appears that little more 
     than 13% of national party spending goes directly into 
     campaigns for media and mobilization activities. Based on our 
     analysis of party spending reports, we believe the bulk of 
     media spending includes the cost of producing and airing 
     television and radio ads. Mobilization spending, in contrast, 
     includes the ``ground'' activity: registering and identifying 
     voters, Get-Out-the-Vote (GOTV) phonebanks and precinct 
     canvassing, and costs of direct mail. Although the national 
     parties spend a small portion of their soft money on these 
     activities, their investments are significant in absolute 
     terms, investing $10.3 million on media-related activities 
     and $7.4 million on mobilization activities.
       State parties use soft money more than other party 
     committees. In the 2000 Elections, they spent 2.5 times as 
     much soft money as national parties. Through transfers, 
     however national parties supply approximately 83% of the soft 
     money that state parties spend for federal related 
     activities. At the state level, 44% of soft money for federal 
     related activities ($149.3 million) is invested in media, a 
     significant increase from the 1996 election in absolute 
     terms, as well as a rise in the portion of the party budget 
     devoted to media. Clearly, state parties are major sponsors 
     of issue ads. Another 12 percent of the budget ($41.8 
     million) goes toward ground mobilization activities, much of 
     it targeted in competitive states. Only 4% of state party 
     budgets reflect grassroots campaign activity that includes 
     distribution of bumper stickers and pins, the staging of 
     rallies and related volunteer work ($11.3 million). State 
     parties rely heavily on soft money for office upkeep and 
     general administrative expenses ($99.5 million or 29% of 
     budget).
       Finally, local parties use very limited amounts of soft 
     money in federal elections. Among the 158 local major parties 
     that submitted campaign finance reports to the Federal 
     Election Commission, their total soft money spending amounted 
     to just $4 million.

 How does soft money spending in the 2000 elections compare to earlier 
                               elections?

       One question that arises in the current debate is whether 
     parties have transformed themselves into campaign media 
     organizations through financing issue ads with soft money. 
     The data provide evidence that state parties have become 
     important venues for producing and airing issue ads, 
     something they did not do prior to the 1996 elections. On the 
     other hand, state parties continue to use soft money for 
     party building activities as they have in the past.
       In the 2000 elections, state parties invested significantly 
     more soft money directly in campaigns than in prior 
     elections. For example, they spent $149.1 million on 
     mediarelated activity, more than double their expenditures in 
     1996. The portion of total party soft money devoted just to 
     media increased from 37% in 1996 to 44% in 2000. Spending on 
     mobilizing voters through the ``ground campaign'' 
     (telephones, canvassing, direct mail) increased from $16 
     million in 1996 to almost $42 million in 2000, a boost of 160 
     percent. The share of the soft money budget devoted to this 
     activity increased from 9 to 12 percent between 1996 and 
     2000.
       In 1998, media and mobilization spending was more evenly 
     distributed than during a presidential election cycle. During 
     the 1998 elections, 17% of soft money went toward media and 
     12% toward ground mobilization. These figures suggest that 
     the media strategies of presidential campaigns drive much of 
     soft money spending. Nonpresidential contests do not always 
     rely as heavily on a media campaign strategy as presidential 
     contests, even though soft money has played an increasingly 
     important role in financing issue ads for congressional 
     campaigns.
       State parties continue to rely a great deal on soft money 
     to maintain the party headquarters, paying for staff 
     salaries, benefits, office equipment and other basic 
     necessities. In the 2000 elections, parties spent almost $100 
     million on administration, a 38% increase from 1996. 
     Administrative costs reflected 29% of all state party soft 
     money spending in 2000, which was a much smaller portion that 
     in the 1996 elections.
       What can we surmise from these data? To the dismay of those 
     seeking definitive evidence to confirm their point of view, 
     the data appear to support both reformers who favor a ban on 
     soft money and those who highlight the virtues of soft money. 
     Advocates of a ban are accurate in observing that the parties 
     abuse their access to soft money by using it for thinly 
     disguised issue ads that actually help the campaigns of 
     particular federal candidates. The parties can hardly claim 
     that their recent media spending is part of a conventional 
     party building strategy when state parties spent virtually no 
     money on issue ads prior to the 1996 elections. On the other 
     hand, those who say a ban on soft money would weaken parties 
     have grounds for concern. It would be wrong to claim that 
     party soft money has not been invested in building the party. 
     Soft money spending on mobilization and grassroots increased 
     substantially with each election for which we have data. It 
     is also clear that soft money pays for a significant share of 
     maintaining the party headquarters.

     What effect will the ``120-Day Rule'' have on party activity?

       The McCain-Feingold bill allows parties to spend soft money 
     up until 120 days before the general election, so long as 
     contributions are capped at $ 10,000 per source. We assume 
     that the motive of this provision is to enable the parties to 
     engage in partybuilding in the early build-up to an election, 
     without letting them use soft money directly in federal 
     campaigns just before the election. With this in mind, we 
     observe how much soft money was spent before and after this 
     120-day marker. We find that only one-quarter of soft money 
     is spent prior to this 120 day marker. Parties invest the 
     vast majority of soft money within the final four months of 
     the election.
       Which activities will be affected the most by the 120-day 
     rule? If the intent of the provision is to root out much of 
     party spending on media activities it might achieve this 
     result. Only 7% of media spending came before the 120-day 
     mark. Of course, under this new rule, parties could simply 
     frontload issue ads

[[Page E263]]

     (if they learn how to craft ads that do not violate other 
     provisions of the new law). But undoubtedly, the impact of 
     media advertising is strongest closer to the election and 
     party strategists will likely seek ways to get around this 
     new provision. Our hunch is that they will spend soft money 
     on issue ads prior to the 120-day marker, and then invest 
     heavily in ``independent'' issue ads that
       In the effort to eliminate soft money issue ads, it appears 
     that party-building activities will also be affected. Only 9 
     percent of spending on voter registration and GOTV activities 
     takes place before the 120-day point. Similarly, only 11% of 
     grassroots and traditional party ``hoopla'' take place before 
     this point. Unsurprisingly, the parties spend significant 
     soft money before the four-month window on maintaining 
     headquarters and raising funds in anticipation of the intense 
     campaign activity to follow. A soft money ban within 120 days 
     of an election will not only reduce party spending on media, 
     but also curtail party building activities Congress intended 
     to encourage through revisions to the Federal Election 
     Campaign Act during the 1970s.

         Are there partisan differences in soft money spending?

       A common concern among policymakers is the relative effect 
     of a ban on either party. Who might be hurt more by banning 
     soft money, Republicans or Democrats? Surely, party members 
     will not want to change campaign finance laws in ways that 
     put their party at a disadvantage. It appears that the 
     Democrats rely more heavily on soft money for direct campaign 
     activity than Republicans. Democrats, for instance, outspent 
     Republicans $87 million to $62 million on media with their 
     soft money. Similarly, Democrats invested more than 
     Republicans in mobilization with soft money, but the 
     difference is not as great as for media. Republicans use more 
     soft money for party overhead than Democrats, and use it 
     slightly more for fundraising.
       The explanation for the Democratic strategy is that the 
     Republicans raise far more hard money than Democrats. It 
     appears, then, that Democrats try to make up for the 
     difference with soft money, using it in ways that might 
     benefit their federal candidates as much as possible. While 
     both parties use soft money to benefit federal candidates 
     directly rather than for generic party building, the 
     Democrats have a far stronger incentive to employ this 
     strategy than Republicans. We can only speculate whether the 
     large figure for ``unidentified'' expenditures ($18.7 
     million) suggests that the Democrats are reluctant to reveal 
     the way they use soft money to influence federal campaigns.
       Given these findings we expect the Democrats to suffer the 
     most from a soft money ban in the short term, since they use 
     it to make up for their relative deficiency of hard money. 
     Over the long-term the Democrats might be able to reach 
     parity with Republicans hard money fundraising, although 
     traditionally the Democrats have been less successful 
     soliciting small contributions than Republicans.
       Table 5 (not shown) is further evidence that soft money is 
     important to both parties in federal elections. It 
     demonstrates that the parties concentrate their money in 
     competitive states. The 10 party organizations that spent the 
     most on media were in states with a highly competitive 
     presidential or Senate campaign, or both. These included 6 
     Democratic and 4 Republican organizations. The average media 
     expenditure among all 100 state parties was 63 cents per 
     voter. Those in the top 10 spent in the range of $1.91 to 
     $9.73 per voter.
       Table 6 (not shown) provides the same analysis for party 
     expenditures on mobilization. The average mobilization 
     expenditure among all 100 state parties was 19 cents per 
     voter. Those in the top 10 spent in the range of 40 cents to 
     $1.3 9 per voter. Interestingly, Democratic organizations 
     comprised the first 8 of 10 organizations in this top 
     category, demonstrating a preference for this mobilization 
     strategy in tightly contested races. For Democratic 
     organizations, the average expenditure on mobilization was 24 
     cents per voter, while it was only 14 cents per voter for 
     Republican committees.


                             Summary Points

       National parties use soft money mostly for party overhead & 
     operations, as well as fundraising. They also transfer 55 
     percent of their soft money to state organizations, which 
     perform much of the campaign work.
       State parties rely on soft money to perform a variety of 
     campaign activities: Approximately 44% was spent on media 
     ($149.4 million); 29% on party overhead and operations ($99.5 
     million); and 15% on direct mobilization and grassroots 
     ($53.1 million).
       The ``120-day rule'' that prohibits soft money spending 
     within 120 days of a general election could eliminate as much 
     as 3/4 of soft money spending: 89% of spending on issue ads 
     falls within 120 days of the general election; and 91% of 
     spending on GOTV and registration falls within 120 days of 
     the election.
       Democrats will likely be hurt by a ban on soft money more 
     than Republicans in the short term: Democrats spend more soft 
     money on media and mobilization than Republicans. Democratic 
     organizations, on average, spent 85 cents on media per voter 
     and 24 cents on mobilization per voter. Republicans, in 
     contrast, spent 42 cents and 14 cents on media and 
     mobilization per voter, respectively.
       The parties concentrate their soft money resources in the 
     closest races: States with competitive presidential contests 
     spent the most on media and mobilization per voter.
       The corrupting influence of unlimited soft money 
     contributions and expenditures, whether real or perceived, is 
     cause for concern and perhaps legislative action. Such action 
     should target underlying problems, while attempting to 
     minimize harmful unintended consequences. The McCain-Feingold 
     bill, with its 120-day amendment and $ 10,000 contribution 
     limit, will eliminate most soft money spending, including 
     spending 12 on thinly disguised candidate ads parading as 
     ``issue ads.'' But it is likely that voter mobilization 
     efforts will be reduced as well. The dramatic increase in 
     soft money media expenditures is driven by the belief that 
     this expensive campaign activity delivers results at the 
     polls. In an effort to prevent corrupt contributions and 
     purge issue ads, the McCain-Feingold bill will constrain the 
     party in other ways. State parties, particularly in states 
     where the parties rely on major donors, will find it more 
     difficult to pay administrative costs, even as they augment 
     efforts to raise money from smaller donors. It is also 
     conceivable that media expenditures will maintain current 
     levels and be paid for with hard money as ``independent'' 
     party expenditures. Given finite resources, broadly based 
     party-building, including voter registration and 
     mobilization, may suffer the most. Certainly there are no 
     guarantees, but it is a plausible outcome that should be kept 
     in mind as the House begins debate on campaign finance 
     reform.

                                  ____
                                  

[Columbia Law Review, April 2000--Symposium: Law and Political Parties]

              * 598 Soft Money, Hard Money, Strong Parties

            (Stephen Ansolabehere and James M. Snyder, Jr.)

       Political parties are central to current efforts to reform 
     campaign finance in the United States. Party money 
     constitutes approximately half of all campaign funds raised 
     at the national level. Limiting party money is, thus, 
     integral to campaign finance reform. This Article examines 
     what might be gained and lost if regulations on party money 
     are imposed. Proponents of stronger (and better financed) 
     parties conjecture that strong parties increase the ability 
     of voters to hold their representative's accountable. We find 
     that such benefits are, in practice, minimal. Instead, we 
     argue that the main benefits of party money, especially soft 
     money, derive from the parties' campaign activities. Soft 
     money finances state party organizations' voter registration 
     and mobilization efforts, which have substantial effects on 
     turnout. Reducing party money will, thus, reduce 
     participation. The benefits of limitations on party soft 
     money must therefore be weighed against likely reductions in 
     voting that would result.


                              Introduction

       American campaign finance law is often described as more 
     loophole than law. Congress and the courts, sometimes working 
     at cross-purposes, continually attempt to clarify and perfect 
     existing regulations, but as campaign practices evolve, 
     candidates, parties, individuals, and groups devise clever, 
     new ways to bend the rules. Today, efforts to reform campaign 
     finance focus on the transfer of national party, funds to 
     state and local organizations. Political parties raise large 
     sums from individuals, corporations, and other associations. 
     They then channel these funds to state and local party 
     organizations, which in turn conduct campaign activities that 
     indirectly and sometimes directly affect federal elections. 
     This was an intended consequence, a genie that Congress meant 
     to let out of the bottle. Our concern is with the effects of 
     putting the genie back in.
       In 1979, Congress amended the Federal Election Campaign Act 
     (``FECA''), [FN1] excluding state and local party building 
     activities from the federal contribution limits. [FN2] The 
     Federal Elections Commission (``FEC'') further clarified the 
     law in a series of rulings, which in essence allow 
     individuals and organizations to give unlimited amounts of 
     money to the national parties' ``nonfederal'' accounts. [FN3] 
     Funds in such accounts are intended * 599 for ``party 
     activities'' at the state and local levels, and may not be 
     contributed to or spent in coordination with federal 
     candidates. Behind this exemption, since termed ``soft 
     money,'' lies a simple objective: Strengthen the political 
     parties.
       The soft money loophole arose in response to two forces: 
     the sorry state of national parties in the 1970s and the 
     long-held belief among political scientists that stronger 
     national parties would improve the ability of voters to hold 
     government accountable. National parties in the U.S. have 
     never had well-financed organizations. [FN4] and in the 
     1970s, their situation appeared especially dire. The national 
     party organizations reputedly needed a greater presence in 
     the new world of campaign finance created by FECA, which put 
     candidates at the center of national political campaigns. 
     [EN5] FECA imposed new restrictions on the amounts that 
     national parties could give to candidates and on the ways 
     that the parties could raise

[[Page E264]]

     money. These restrictions hit the Democratic National 
     Committee (``DNC'') particularly hard, as the committee 
     labored under debt from the 1968 and 1972 campaigns.
       Soft money also answered a century of political science 
     speculation and theorizing. Political scientists have long 
     argued that the absence of strong national party 
     organizations in the U.S. limits the ability of voters to 
     hold government accountable for public policies. [FN6] In 
     order to hold the government accountable, voters need to face 
     clear, programmatic choices. Party money is one means to this 
     end.
       Parties are able to impose discipline on their members, 
     acting as a counterweight to the many special interests that 
     may chip away at the public good behind legislation. [FN7] 
     Expanding party money might also weaken the influence of 
     interest * 600 groups by lessening the unique campaign 
     finance advantages of incumbents, which derive substantially 
     from interest group contributions. [FN8] This argument 
     received its most famous expression in a report of the 
     American Political Science Association published in 1950 and 
     entitled ``Toward a More Responsible Two-Party System.'' 
     [FN9]
       The committee that crafted the report recommended three 
     concrete changes in the practice of politics that would 
     improve accountability: more programmatic parties, greater 
     democracy within the parties (such as primary elections), and 
     deregulation of the parties' campaign fundraising activities. 
     [FN10] On all three counts, American politics have moved in 
     the direction of the committee's proposals and its vision of 
     responsible party government. The parties within Congress 
     exhibit much more party line voting today than they did in 
     the 1970s, providing voters a much clearer choice. [FN11] 
     Primary elections and other party reforms in the 1960s and 
     1970s created more democracy within the parties. [FN12] The 
     consequence of those changes in our national politics is the 
     subject of intense scholarly scrutiny. [FN13] Our concern is 
     with the third factor in the contemporary experiment with 
     stronger parties, the money.
       Today, the American political parties are prolific 
     fundraisers. In 1998, for example, Democratic and Republican 
     national party organizations raised $445 million for their 
     federal (hard money) accounts, and $224 million for their 
     non-federal (soft money) accounts. [FN14]
       Party money has not been heartily embraced by the public, 
     politicians or political scientists. Twice, Congress has 
     nearly closed the soft money loophole. [FN15] Campaign reform 
     bills proposed by Representatives Shays and Meehan and by 
     Senators McCain and Feingold passed the House of 
     Representatives in each of the last two Congresses, and have 
     failed in the Senate only because a minority of senators 
     sustained filibusters. *601 [FN16] Scholarly and popular 
     commentary has similarly turned against party finance, and 
     against soft money in particular. The objections are not that 
     the parties have become too strong. Rather, it is alleged 
     that party finance practices have inadvertently increased the 
     political leverage of interest groups [FN17] and have ruined 
     the ability of government agencies to regulate the system of 
     political finance. [FN18]
       Specific objections to soft money emphasize the evasion of 
     existing limits. Following Buckley v. Valeo, [FN19] 
     contribution limits on individuals and groups became the 
     centerpiece of campaign finance regulations in the United 
     States. [FN20] In each election, individuals may give no more 
     than $1,000 to a candidate, up to $20,000 to national party 
     organizations, and a total of no more than $25,000 to all 
     federal candidates. Organizations may give no more than 
     $5,000 to a candidate and $15,000 to national party 
     organizations in each election. The 1979 amendments to FECA 
     provide an avenue through which groups and individuals can 
     avoid these limits, giving unlimited amounts to the parties' 
     non-federal accounts. [FN21] In addition, organizations, 
     especially corporations, can avoid having to set up a 
     separate and segregated fund, commonly called a political 
     action committee (``PAC''), through which the organizations 
     raise money for federal elections. Finally, soft money is 
     widely seen as an evasion of presidential spending limits, as 
     presidential candidates can raise money for the ``non-
     federal'' accounts of the DNC and RNC and those funds can be 
     spent in battleground states.
       Critics of contemporary campaign finance raise a more 
     generic concern about party money. Interest groups might 
     capture or corrupt the parties, just as they allegedly 
     compromise congressional decisionmaking. Soft money is raised 
     without contribution limits; many of the donations exceed 
     $100,000 and come from corporations, associations, and 
     individuals with strong interests in legislative and 
     executive decisions facing the government. Large donations 
     from a specific interest or industry, it is feared, might 
     convince the party caucuses within Congress or the president 
     to protect that interest. Our aim is to put the essential 
     claims about party money to empirical scrutiny. First, how 
     apt is the traditional view of parties? Does party money 
     produce greater degrees of electoral accountability and 
     legislative discipline? Second, how accurate are contemporary 
     critics of parties? Is party money, especially soft money, 
     swamping the system? Has FECA's system of contribution limits 
     broken down? Finally, what would be the *602 practical 
     political consequences of further constraining party money 
     raised for state and local parties and elections? We analyze 
     these questions through the lenses of campaign finance 
     reforms that would end soft money.
       Part I of this paper details how parties raise money and 
     how they handle it, especially in contrast to how candidates 
     raise money. Here we assess how much money would be affected 
     by proposals to close the soft money loophole, and whether 
     the rise of soft money signals the failure of the 
     contribution limits established by FECA. Part 11 assesses two 
     key claims about party money and party discipline in national 
     politics: that party contributions and expenditures foster 
     electoral competition and that party money creates greater 
     party discipline within the legislature. In this Part we 
     argue that the parties do produce more electoral competition, 
     but that a ban on soft money would have little impact on 
     national party politics. Part III of this paper explores how 
     party money is used at the state level, especially for grass 
     roots activities. Here we project that a complete ban on soft 
     money would significantly curtail grassroots activities of 
     state party organizations and would significantly reduce 
     participation, a consequence that has as yet received little 
     attention in national discussions about campaign finance 
     reform.

                  I. Party Money Amounts and Accounts

       Party money is extremely important: It accounts for nearly 
     half of all campaign money raised at the
       To measure the importance of national party fund raising we 
     contrast the resources of parties to those of candidates, 
     rather than to interest groups. There are four reasons for 
     this contrast. First, recent political science scholarship 
     emphasizes that candidates have eclipsed parties as an 
     organizing force in voting behavior and elections. [FN23] The 
     rise of personal voting and the incumbency advantage over the 
     last forty years suggests that many voters focus on the 
     individual politician more and rely on party less. [FN24] 
     Individuals and groups may choose to channel their resources 
     either to parties or individual candidates. The decline of 
     parties is often traced to increases in the campaign 
     resources of individual candidates, especially incumbents, 
     and declines in electoral resources and activities *603 of 
     party organizations. [FN25] Do parties command substantially 
     fewer resources than candidates do? Surprisingly, in our 
     world of candidate-centered campaigns, parties and candidates 
     attract approximately the same amount of money. [FN26]
       Second, party committees are essentially campaign 
     operations and are, therefore, most appropriately compared to 
     candidates, rather than to political action committees. 
     Parties seek to win a majority of seats in the legislatures 
     or control of the executive office, and they do so through 
     direct campaigning and by assisting their local campaign 
     organizations. FEC reports reveal that parties are not 
     primarily operations for donating money to candidates. Less 
     than one percent of party money is contributed to federal 
     candidates, and only about ten percent is spent on their 
     behalf. [FN27] FEC audits of party committees reveal that the 
     national Republican and Democratic organizations spend their 
     funds on overhead, fundraising, and their own campaigns to 
     win control over government, including grassroots organizing 
     and television advertising, as well as on recruiting and 
     training candidates and campaign organizers. An audit of the 
     RNC's 1984 accounts revealed that approximately thirty 
     percent of that money was spent on direct campaign 
     activities, such as advertising and field operations; an 
     additional third went to fundraising. [FN28] An audit of the 
     Dukakis campaign in California, and of the California State 
     Democratic party, in 1988 showed that half of the funds went 
     for field operations, such as get-out-the-vote drives, 
     canvassing, and direct mail; twenty percent went for media. 
     [FN29] These figures are remarkably similar to the activities 
     of federal candidates, whose reports to the FEC reveal that 
     thirty percent of congressional campaign money goes for media 
     advertising and about twenty percent goes for grassroots 
     cwnpaign activity. [FN30]
       Third, candidates are the relevant benchmarks against which 
     to compare party finance because the government is ultimately 
     organized both by individual politicians and by parties of 
     politicians. The U.S. House and Senate, for example, are 
     organized into committees, which are often tailored to 
     members' and constituents' interests, as well as party 
     hierarchies. If contributors can influence public policy or 
     change the composition *604 of the government with their 
     campaign donations, then they may be able to achieve their 
     ends either through donations to politicians or to party 
     committees. Finally, parties and candidates draw on the same 
     pools of donors--individuals and organizations, especially 
     corporations. But, different restrictions apply to candidate 
     and party fund raising. Donations to federal candidates and 
     party committees fall under the contribution limits imposed 
     by FECA, and individuals and groups face lower contribution 
     limits when they give to candidates than when they give to 
     parties. Also, non-federal party accounts are not subject to 
     federal contribution limits. [FN31]
       Table 1 contrasts the receipts of federal candidates and of 
     national party committees over nine election cycles in the 
     1980s and 1990s. The first and second columns of the

[[Page E265]]

     table display the amounts raised by congressional (House and 
     Senate) and presidential candidates, respectively. The 
     Presidential funds include public funds provided in the 
     primary and general elections. The third and fourth columns 
     display the parties' federal (hard money) and non-federal 
     (soft money) receipts. [FN32] Official FEC reports on soft 
     money are not available for the early 1980s, and the amounts 
     are presumed to be small. For 1984 and 1988, we include some 
     estimates provided by the Citizens' Research Foundation and 
     other sources.

                   Table 1. Candidate and Party Funds

       Some double counting exists in the table because party 
     committees and candidates may contribute money to each other. 
     Surprisingly little money flows between parties' and 
     candidates' treasuries. Parties contributed only about $5 
     million directly to candidates in 1998, accounting for less 
     than one percent of candidates' funds. [FN33] In addition, 
     parties' coordinated and independent expenditures totaled $67 
     million in 1998. In total, parties only spent about ten 
     percent of their funds on federal election campaigns at the 
     national level, and almost all of these expenditures are 
     advertising and other coordinated and independent 
     expenditures made by the parties, rather than direct 
     contributions to candidates. Even less money goes from 
     candidates to parties. Federal candidates contribute only a 
     trace of their money to national party committees. [FN34] 
     Parties and candidates, then, represent distinct campaign 
     fund raising venues in national politics; one does not feed 
     the other. More money flows among the party committees. 
     Twelve percent of soft money in 1994 and ten
       The contrast between party and candidate national 
     fundraising helps put the parties in clearer relief. First, 
     the data in Table 1 reveal that under FECA candidates and 
     parties play roughly equal roles in campaign *606 fund 
     raising. Federal and non-federal money, amounted to 
     approximately forty percent of all money raised at the 
     national level in the 1990s. [FN36] The fraction of all money 
     going to candidates in Table 1 is fairly stable, averaging 60 
     percent, but never more than 65 percent or less than 55 
     percent. The equality of candidate and party money is 
     somewhat surprising given the emphasis within political 
     science on the rise of ``candidate-centered'' campaigns. The 
     importance of parties, though, should not be seen as evidence 
     that FECA has gradually broken down. The parties appear to be 
     a constant in American campaign finance, and this is 
     accommodated by FECA.
       Second, the parties have changed how they handle their 
     funds somewhat, relying increasingly on non-federal accounts. 
     Much of the recent growth in the parties' treasuries has come 
     through non-federal funds. The hard money accounts of parties 
     have grown much more slowly than the hard money accounts of 
     candidates. From the early 1980s to the late 1990s. House and 
     Senate money (in off-year elections) grew 85 percent. Over 
     that same time period, the federal accounts of the parties 
     grew by only 45 percent. [FN37] Parties have kept pace with 
     candidate fund raising through soft money.
       Reports on the amounts of soft money before 1992 are highly 
     incomplete. The Citizens Research Foundation estimated the 
     amount of soft money in 1984 as $22 million, which amounts to 
     about 5 percent of the party money raised that year. [FN38] 
     The figures in Table 1 reveal that by the end of the 1990s 
     soft money had risen to $200 million.
       Third, soft money, though it has grown, is still a 
     relatively small fraction of party money and of all money. By 
     the end of the 1990s, non-federal accounts handled slightly 
     less than a third of all party money. Even with this growth, 
     non-federal accounts still handle much less money than 
     federal accounts, and contributions to and expenditures from 
     federal accounts must still comply with contribution limits 
     set by FECA. Table 1 reveals that by the end of the 1990s, 
     soft money accounted for just 15 percent of all money.
       Soft money, as it is commonly discussed, has a more 
     negative connotation than simply the amounts of money flowing 
     into non-federal accounts. Soft money has become synonymous 
     with money laundering. Parties, interest groups, and 
     candidates reputedly avoid the limits on group and individual 
     contributions to federal candidates and parties by funneling 
     money raised in national accounts to state and local 
     organizations. The state and local organizations serve merely 
     as fronts for the federal candidates' and parties' campaigns. 
     How much money exceeds the limits?
       Total non-federal party money provides an upper bound 
     estimate of the amount of money that is given in order to 
     evade contribution limits *607 on individuals, corporations, 
     and other associations. The figures in Table 1 suggest that 
     the amount evading the contribution limits is small relative 
     to the amounts subject to the limits. Even with the soft 
     money loophole, two-thirds of all contributions to parties go 
     to federal accounts and are subject to contribution limits. 
     Party money constituted about 45 percent of national campaign 
     finance in 1998; non-federal accounts handled only 15 percent 
     of all money raised in 1998 at the national level.
       Not all donors to non-federal accounts exceeded the limit 
     that they would have been subject to had they contributed to 
     a federal committee. In 1998, approximately 18,000 different 
     donors gave to the national parties' soft money accounts. 
     Only eleven percent gave more than $20,000 to soft money 
     accounts, which is the limit on contributions to party 
     committees. This relatively small number of donors gave 78 
     percent of the soft money--that is, $153 million of $196 
     million. This is a very large amount of money ``skirting'' 
     the limits, and it is a cause for concern. However, it 
     represents only 12 percent of all money raised by candidates 
     and parties at the national level. [FN39]
       Closing the soft money loophole will not force all of the 
     money in parties' non-federal accounts out of politics. A 
     sizable amount of non-federal money ($42.6 million out of 
     $196 million) is raised within contribution limits, and we 
     suspect that federal committees would likely attract these 
     funds if non-federal accounts did not exist. [FN40] In 
     addition, if soft money is banned, donors to these accounts 
     might redirect their contributions to other accounts--to hard 
     money accounts, to candidate accounts, or to state and local 
     organizations directly.

                 II. Party Money in National Elections

       Current campaign finance reform efforts, such as the 
     McCain-Feingold bill, aim to eliminate soft money entirely. 
     If those efforts succeed, what will be the consequences for 
     national and state politics? We turn first to the national 
     level. The concern for national politics, as political 
     scientists have described it, is whether restrictions on 
     party resources would lessen the ability of voters to hold 
     the government collectively accountable.
       A further concern is how the parties conduct their own 
     campaigns. If a sizable share of the money goes to voter 
     registration and mobilization, then party money might foster 
     accountability by encouraging people to vote and make their 
     preferences heard. Here, we address the first two concerns; 
     we leave the third issue to the next section, as that is more 
     readily addressed at the state and local level.
       A. Does Party Money Increase Electoral Competition?
       Many students and observers of Congress complain bitterly 
     about the lack of competition in congressional elections. 
     They cite such facts as the high incumbent reelection rate 
     (averaging over 95 percent since 1980): [FN43] the 
     ``vanishing marginals''; [FN44] the incumbency advantage in 
     voteshare, around 8 percent: [FN45] and the huge advantage 
     incumbents have in fundraising. [FN46] Weakening the 
     fundraising capabilities of parties would probably reduce 
     competition. [FN47] Party money flows to more competitive 
     races. [FN48] Also, as we show here, party money flows much 
     more freely to non-incumbents than PAC money does. [FN49] The 
     panels on the left-hand side of Figure I show the natural log 
     of party money (including coordinated expenditures), plotted 
     against the Democratic vote-share, for each House race 
     between 1978 and 1998 that was contested by both major 
     parties. The top left panel shows the relationship between 
     Democratic party money *609 and electoral closeness: the 
     bottom left panel shows the relationship between Republican 
     party money and electoral closeness. A unit change in the 
     logarithmic scale can roughly be interpreted as a one percent 
     change in the variable. The graphs, then, represent how a 
     percentage point change in the Democratic vote share 
     corresponds to a percentage chance in the amount of party and 
     PAC money received. The symbols are ``I'' for races 
     incumbents. ``C'' for races with challengers, and ``0'' for 
     open seat races.
       Parties clearly, target closer races. The curves on the 
     left have a distinct inverted-U shape, with a peak at almost 
     exactly .5. showing that Democratic and Republican party 
     committees target close races. Also, the curves are 
     relatively symmetric at about .5, suggesting that the party 
     committees concentrate equally on vulnerable incumbents and 
     credible challengers, and tend to ignore safe incumbents and 
     struggling challengers.

[[Page E266]]

 Figure 1: Party versus PAC Contributions as a Function of Democrat's 
        Share of the Vote Won, US House Elections, 1978 to 1998.

       The panels on the right-hand side show analogous plots for 
     PAC contributions. Although PACs also tend to target close 
     races, there is one striking difference between PACs and 
     parties. While party contributions drop off sharply in 
     noncompetitive races involving incumbents, PAC contributions 
     do not. PACs give nearly as much to safe incumbents as they 
     give to incumbents who are in trouble. Incumbency of course, 
     is nearly synonymous with victory, as approximately 95 
     percent of incumbents who *610 seek reelection win. [FN50] 
     The asymmetry suggests that PACs are drawn to candidates who 
     are more likely to win. Elsewhere, we have shown that this 
     behavior is consistent with the argument that interest groups 
     give money as an investment in politics, with some 
     expectation of a return for their donation. [FN51] Also, 
     though this is more difficult to discern from the graph, PACs 
     give significantly more to incumbents than to non-incumbents, 
     holding the vote margin constant.
       Specifically, PACs give more money to incumbents than they 
     do to open seat candidates who win by the same vote margin or 
     who compete in districts with similar partisan levels. [FN52] 
     What is more, PACs give more to incumbents who lose by I to 5 
     percent than they do to challengers who win by I to 5 
     percent.[FN53] This is not true for the parties. Party 
     contributions act as something of a counterbalance to PAC 
     contributions. A back-of-the-envelope calculation gives some 
     sense of what might happen if party money dries up. Consider 
     all races between 1988 and 1998. On average, Democratic 
     challengers received 14 percent of their total campaign funds 
     (including coordinated and independent expenditures) from 
     party committees, and Republican challengers received almost 
     11 percent of their funds from parties. The corresponding 
     figures for incumbents were 2.0 percent and 1.7 percent, 
     respectively. In a previous paper, we estimated that the 
     elasticity of challenger vote-share with respect to 
     challenger spending .05 to .08 range. [FN54] Using an 
     intermediate value of .07, a 10 percent reduction in 
     challenger spending implies that the average challenger's 
     vote percent will fall by about 2.5
       The effect of this counterweight of parties in specific 
     races for Congress is slight, primarily because parties spend 
     so little money on individual races. We are unsure what the 
     effects on electoral competition and turnover might be if the 
     parties were to spend. say, five times more than they 
     currently do on national elections. This is the world 
     envisioned by proponents of stronger parties. such as Dan 
     Lowenstein, who recommends heavy public subsidy of parties to 
     counteract the incumbency advantage. [FN55] The obstacle to 
     forecasting what this world would be like is that it is 
     unclear what circumstances would lead the parties to shift 
     their *611 resources more heavily into congressional 
     campaigns and away from state and local activities.
       The bottom line, though, is that significant reductions in 
     party receipts would not change competition in the national 
     elections appreciably. Complete elimination of party 
     contributions and coordinated and independent expenditures 
     would lower challengers' vote shares by 2.5 percent, but the 
     typical challenger today only receives 35 percent of the 
     vote.

          B. Does Party Money Buy ``Loyalty'' to the Parties?

       One way to buy loyalty is to help elect and reelect those 
     who are known to be loyal. There is some evidence that at 
     least for Democrats, party committees give more money to 
     House members who vote in line with their party's leaders. 
     [FN56] The evidence is rather weak, however, and other 
     studies find no effects. [FN57]
       Showing that party money actually affects voting records is 
     even trickier, because it is extremely difficult to control 
     for the ``baseline'' level of party support (how much a 
     member would support the party even if he or she did not 
     receive party money). Leyden and Borrelli claim to show an 
     effect, but it is doubtful that have correctly controlled for 
     the baseline. [FN58]
       We find mixed evidence for the first claim. We ran a series 
     of Tobit regressions predicting party money as a function of 
     electoral circumstances and party loyalty in roll call 
     voting, measured as proximity to the parties' medians. [FN59] 
     Using the estimated relationship we can measure the *612 
     expected amount of party money received by loyal and maverick 
     incumbents. Contrast a Democrat who is at the party's median 
     with one at the 25th percentile of his party (in the 
     conservative or moderate direction). The average Democratic 
     incumbent over the period 1978 to 1998 received about $10,000 
     from the Democratic party's committees. [FN60] The more 
     moderate Democrat received only about $6,000 from party 
     committees. In other words, the effect of being at the 25th 
     percentile of the Democratic party, rather than at its 
     median, cost the more moderate member about $4,000 in party 
     campaign funds. For Republicans, the corresponding difference 
     between the party's median member and a member at the 25th 
     percentile (again in the moderate direction) is just $206--
     essentially no effect. [FN61]
       These slight effects suggest that further restrictions on 
     federal party contributions and spending money would have 
     relatively little effect on discipline within the party. 
     Nevertheless, these effects measure how we predict loyalty 
     rates to change with modest changes in party contributions. 
     It is unclear what the consequences for party discipline 
     might be if the party committees' presence in candidates' 
     campaigns expanded significantly. What seems more certain is 
     that expanding the parties' campaign activities and 
     expenditures would aid challengers somewhat. Parties 
     contribute and spend money in federal races in ways that 
     foster competition. The sums, however, do not appear large 
     enough to make an appreciable difference in the final 
     election outcomes. If anything, the behavior of the parties 
     in national elections suggests that, if our objective is to 
     increase electoral competition and, thus, electoral 
     accountability, then parties command too little of the money 
     spent in American national elections.
       One caveat to this implication is in order. Party money is 
     not politically balanced or neutral: Republican committees 
     regularly raise and spend more money than Democratic 
     committees. This pattern is especially strong in hard money 
     accounts; in 1998. Republican committees raised $285 million 
     and Democratic committees raised $160 million in hard money. 
     Over the last decade (1988 to 1998), Republican national 
     campaign committees raised 65 percent of the party money in 
     federal accounts. [FN62] Soft money is more balanced: Over 
     the last decade, Republican *613 committees have accounted 
     for 55 percent of soft money. [FN63] Thus, complete 
     deregulation of party money would likely benefit Republican 
     committees and candidates, and elimination of the soft money 
     loophole may benefit the Democrats.

             III. Party Money in State and Local Elections

       Closing the soft money loophole would affect state and 
     local party organizations and the voters they reach much more 
     acutely than it would affect national politics. The effects 
     would be two-fold. First, as we show here, eliminating soft 
     money would seriously reduce the party treasuries in many 
     states. Second, eliminating soft money will significantly 
     reduce the campaign activities that state and local party 
     organizations conduct. Soft money appears to subsidize a wide 
     range of activities, including get-out-the-vote drives, 
     broadcast advertising, and day-to-day operations of the 
     organizations. Of
       To provide a thorough accounting of state parties' 
     financial activities, we chose to profile three states--
     Idaho, North Carolina, and Ohio--across three election 
     cycles, 1991-92, 1993-94, and 1995-96. [FN64] These states 
     appear representative of the rest of the country. Ohio is the 
     seventh most populous state in the U.S., and it has highly 
     competitive elections. [FN65] North Carolina is a mid-sized 
     state (two-thirds the population of Ohio); it too is highly 
     competitive. [FN66] Idaho is a small state, and leans 
     strongly toward the Republicans, though Democrats have won 
     statewide and federal offices over the last two decades. 
     [FN67]
       These states share some important characteristics. All 
     three states have very complete public reporting of the 
     receipts and expenditures of the parties. [FN68] In all three 
     states, the parties' central or executive committees handle 
     almost all of the parties' campaign money. Idaho's party 
     committees raised a total of $3 million in 1996, $2.4 million 
     of which *614 went to the state committees--the Idaho 
     Democratic Party and the Idaho Republican Party. The 
     remainder was distributed evenly across numerous county 
     party, committees. [FN69] The House and Senate caucus 
     committees controlled relatively little. North Carolina also 
     has extremely active county committees; they tend to be 
     recipients of state party money. [FN70] In Ohio, party money 
     is concentrated in the state committees, though the 
     legislative caucuses have played a relatively more important 
     role in the past. [FN71] In terms of party money, these 
     states span much of the observed variation in transfers from 
     non-federal accounts. Combining the 1996 and 1998 elections, 
     Ohio, with 11 million people, received $10.6 million in soft 
     money, North Carolina, with 7.5 million people, received $7.6 
     million in soft money, and Idaho, with 1.2 million people, 
     received $2.4 million in soft money. [FN72]

      Figure 2: Repub. and Democ. Soft Money in States, 1996-1998

       Tabular or graphic material set forth at this point is not 
     displayable.
       Figure 2 graphs the combined 1996 and 1998 soft money 
     contributions to state parties of Democratic and Republican 
     national committees. [*615 FN73] The graph shows that 
     Democratic and Republican money increase together. In the 
     figure, California, with 32.7 million people, receives by far 
     the most money ($33 million). Ohio ranks with New York and 
     Illinois as the next largest recipients of soft money over 
     the last two election cycles. North Carolina is in a cluster 
     of states that includes Texas, Pennsylvania, Washington, and 
     Kentucky. Idaho is in a cluster of smaller states including 
     Arkansas,

[[Page E267]]

     Connecticut, and Massachusetts, which received about $2 
     million total over the two election cycles. The densest 
     cluster of states in the graph is in the lower left corner. 
     One-third of the states received less than $1 million from 
     national parties non-federal accounts over the two elections.
       The importance of soft money to state parties is readily 
     measured as the fraction of state party committees' total 
     receipts that come from non-federal national party accounts. 
     How dependent are the states on non-federal accounts for 
     their resources? Using the reports of the state election 
     finance agencies in Idaho, North Carolina, and Ohio, we 
     calculated the total amount of money raised by the party 
     committees in these states, excluding transfers between state 
     party committees. In Idaho, the Republican and Democratic 
     state and local party committees raised a combined total of 
     $1.4 million in 1992, of which $550,000 was soft, and a total 
     of $3.0 million in 1996, of which $1 million was soft. In 
     North Carolina, the state and local party committees raised a 
     total of $10 million in 1992, of which $4.9 million was soft, 
     and a total of $18 million, $5.7 million of which came from 
     soft money accounts. In Ohio, the state and local party 
     committees amassed total receipts of $24.7 million in 1992, 
     of which $8.1 million was soft, and $19.6 million in 1996, of 
     which $7.6 million was soft.
       These figures show that state parties depend heavily on 
     soft money transferred from the national party committees. In 
     all of these cases, more than one-third of the state 
     organizations' total funds came from national ``non-federal'' 
     accounts. Cutting soft money would significantly reduce state 
     parties' financial resources.
       How would reduction of these funds affect state parties' 
     activities? It is often charged that non-federal money merely 
     takes the form of advertising for federal candidates cloaked 
     as state and local party building. This perception appears to 
     be wrong for three reasons.
       First, national party committees (such as the DNC-State 
     Account) contribute or transfer money directly to the state 
     party committees rather than spend money in the states, which 
     the national committees might do if they were advertising for 
     federal candidates. Nor is the money that they do spend 
     clearly earmarked. Instead, almost all of the soft money that 
     flows into these states is transferred to the general 
     treasuries of state committees, *616 which are controlled by 
     state party organizations. The ultimate decision about how 
     the money is to be spent, it seems, rests with the state 
     committees, rather than with federal committees or federal 
     candidates.
       Second, to the extent that we observe direct national 
     expenditures in the states or earmarked money, these funds 
     are dedicated to overhead, such as office expenditures. Ohio 
     is the clearest example. In 1992, the RNC and DNC spent 
     almost ten million dollars in the state; eighty percent of 
     these funds
       Third, the state party organizations spend considerable 
     sums on field or grassroots campaigning, such as direct mail, 
     precinct walks, and voter registration. Documenting the 
     amounts spent on various activities takes considerable 
     effort. Working with the public reports of the parties filed 
     with the state elections commissions in Idaho, North 
     Carolina, and Ohio, we classified each itemized expenditure 
     by the parties in the 1991-92 and 1995-96 election cycles. 
     [FN76] All told there were over 41,000 separate data entries 
     to classify. We divided the expenditures into a fairly 
     detailed category scheme that paralleled the format developed 
     by Dwight Morris and his collaborators. [FN77] We then 
     aggregated these into several broader categories: 
     ``grassroots or direct campaigning,'' ``media campaigning,'' 
     ``overhead,'' ``consulting,'' ``contributions,'' 
     ``transfers,'' and ``fundraising.''
       The state parties spend fairly constant proportions on each 
     of these categories. For our purposes grassroots and media 
     campaigning are of greatest interest. In North Carolina, in 
     both 1992 and 1996, we estimate that the state parties spent 
     approximately 20 percent of their funds on media advertising 
     and 25 percent on grassroots or direct campaigning. In Idaho, 
     we estimate that the parties spent approximately, 15 percent 
     of their funds on grassroots campaigning in 1992 and 1996. 
     They spent just 4 percent on media in 1992 and 9 percent on 
     media in 1996. In Ohio, we estimate that the parties spent 7 
     percent on media in 1992 and 5 percent on media in 1996. 
     They, spent 32 percent on grassroots campaigning in 1992 and 
     27 percent on grassroots campaigning in 1996. In each of 
     these states, the parties spent between 30 and 40 percent of 
     their funds on reaching voters directly, the larger category 
     of voter contact being direct voter contact, such as direct 
     mail and canvassing, not broadcast advertising. [FN78]
       How will the elimination of party soft money affect the 
     campaign activities that parties conduct? Between 1992 and 
     1996 we observe for each state changes in the total receipts 
     of the party treasuries as well as changes in their 
     grassroots, or direct campaign, expenditures. The ratio *617 
     of the change in grassroots expenditures to the change in 
     total receipts measures how the parties translate marginal 
     changes in their receipts into changes in their activities. 
     From 1992 to 1996, Idaho parties' receipts rose $1.5 million. 
     Their expenditures on direct voter contact rose $250,000. For 
     every additional dollar raised, the Idaho parties spent an 
     additional 16 cents on voter contact. From 1992 to 1996, 
     North Carolina parties' receipts rose $8.4 million. Their 
     expenditures on direct voter contact rose $2.15 million. For 
     every additional dollar raised, the North Carolina parties 
     spent an additional 25 cents on voter contact. From 1992 to 
     1996, Ohio parties' receipts fell $7 million. Their 
     expenditures on direct voter contact shrank, $1.3 million. 
     For every dollar lost, the Ohio parties reduced expenditures 
     on voter contact 18 cents. [FN79] These figures suggest that 
     every dollar lost by the parties from a reduction in federal 
     transfers would cut expenditures on state parties' direct 
     campaign activities by 20 cents. [FN80] Soft money transfers 
     to these states totaled $13 million in 1996. [FN81] 
     Elimination of these funds, we estimate, would cut the state 
     parties direct campaign expenditures by $2.6 million dollars.
       How much would turnout decline? In 1996, 7.5 million people 
     voted in Idaho, North Carolina, and Ohio combined. [FN82] To 
     calculate how many fewer people would have turned out without 
     the soft money subsidy of state grassroots activities, we 
     need to know the cost of getting an additional voter to the 
     polls through these activities. From a series of ingenious 
     field experiments. Alan Gerber and Donald Green have 
     estimated the marginal cost of getting an additional person 
     to the polls through canvassing and related means of voter 
     contact. [FN83] They estimate that mobilizing an additional 
     voter costs between $15 and $20. [FN84] These figures suggest 
     that between 170,000 and 130,000 fewer people would have 
     voted in these states in 1996 without the grassroots 
     activities underwritten by the national parties' soft money. 
     In other words, cutting soft money would have lowered turnout 
     in these states by slightly more than two percentage points. 
     [FN85]


                            *618 Conclusion

       Party money poses a dilemma, both for those who advocate 
     stronger and more responsible parties and for those who 
     advocate elimination of soft money to reform campaign 
     finance. Broadly speaking, political parties are thought to 
     be instruments of greater political accountability and mass 
     democracy.
       Voters can more readily hold stronger, unified national 
     parties responsible for their actions and redirect government 
     if need be. At least since the 1950s, political scientists 
     have argued that we should strengthen the parties 
     organizationally, and that unregulated party campaign money 
     is one of the main mechanisms through which the United States 
     can achieve stronger parties. [FN86] The devil, though, is in 
     the fund-raising. The parties may have to act irresponsibly 
     toward the public in order to raise funds from wealthy 
     individuals, corporations, and other private concerns.
       We have considered the concrete tradeoffs presented by 
     proposals to eliminate soft money. In terms of reducing 
     corruption or undue influence, such proposals, at best, can 
     eliminate money that exceeds existing limits. We estimate 
     that soft money contributions that evade existing 
     contribution limits amounted to approximately $150 million in 
     the 1998 elections. We are unsure what that money buys.
       We know of no research that provides reliable estimates of 
     the amount of influence purchased with
       Individual legislators do not depend on these funds at 
     all--if anything these funds are a nuisance, as the 
     expenditures likely go to support their opponents. Thus, if 
     the effects are corrupting, they are not corrupting of 
     individual legislators. Soft money might, however, unduly 
     influence the parties. Even still, the national parties raise 
     two-thirds of their money in hard money donations.
       Against these possible consequences must be weighed the 
     possible effects of soft money on the ``responsibility'' of 
     the parties. We have focused on three of the central 
     arguments about what responsibility means: party discipline 
     in policy-making, national electoral competition, and party 
     building activities, especially grassroots mobilizing.
       We believe that the effects of eliminating soft money on 
     the ability of the parties to present voters with clear, 
     programmatic choices would be slight. Contrary to the 
     responsible party argument, party money evidently, does not 
     correspond with significantly more party discipline. *619 
     Among Democrats party, loyalty within Congress and party 
     contributions are correlated; among Republicans, they are 
     not. And the sums are so small that it seems unlikely that 
     the parties have created greater discipline through their 
     campaign finance committees.
       Party money, if it continues to grow, might have 
     substantial consequences for national elections. In 
     particular, party money has the potential to counterbalance 
     interest group contributions in congressional elections. PACs 
     account for most of the incumbency advantage in campaign 
     finance. [FN87] Parties, by contrast, give to close races, as 
     suggested in Figure 2, and spend their money

[[Page E268]]

     efficiently so as to have the largest effect on electoral 
     outcomes. More party money in congressional elections, at 
     least relative to interest group money, would probably 
     produce much higher electoral competition. However, the 
     parties currently give little to congressional candidates and 
     spend little on individual races.
       The most troubling effect of closing the soft money 
     loophole is that it would significantly lessen the electoral 
     presence of state and local party organizations. Debates in 
     Washington on bills designed to eliminate soft money, and 
     many political science and popular journals have discussed 
     the many ramifications of eliminating soft money. Little 
     mention, however, has been made of the consequences for the 
     state parties and the voters that they reach. Closing the 
     loophole will starve many grassroots activities of state and 
     local parties. Eliminating all current soft money 
     expenditures, we estimate, would lead to a 2 percent decline 
     in voter turnout--without soft money, approximately 2 million 
     fewer Americans would have gone to the polls in 1996.


                                ENDNOTES

     [FN1]. 2 U.S.C. Sec. Sec. 431-55 (1994).
     [FN2]. See Elizabeth Drew, Politics and Money: The New Road 
     to Corruption 15 (1983); Frank J. Sorauf, Inside Campaign 
     Finance 149 (1992).
     [FN3]. There are numerous FEC advisory opinions that clarify 
     what does and does not constitute a non-federal expenditure 
     or contribution. Three are of particular note: Allocation of 
     Costs for Voter Registration. [1976-1990 Transfer Binder] 
     Fed. Election Camp. Fin. Guide (CCH) P 5340 (FEC Advisor 
     Opinion 1978-10); Corporate Support for Party Convention, 
     [1976-1990 Transfer Binder] Fed. Election Camp. Fin. Guide 
     (CCH) P 5348 (FEC Advisory Opinion 1978-46); Get-Out-the-Vote 
     Drive for State Candidates, [1976-1990 Transfer Binder] Fed. 
     Election Camp. Fin. Guide (CCH) P 5353 (FEC Advisory Opinion 
     1978-50). See also Richard Briffault, The Political Parties 
     and Campaign Finance Reform, 100 Colum. L. Rev. 620, 629 
     (2000) (discussing the FEC's role in clarifying non-federal 
     party activities).
     [FN4]. See American Pol. Sci. Ass'n, Toward a More 
     Responsible Two-Party System, 44 Am. Pol. Sci. Rev. 45 (Supp. 
     1950).
     [FN5]. See Sorauf, supra note 2, at 28-42.
     [FN6]. The intellectual history on this is long and storied. 
     See, e.g., Austin Ranney, The Doctrine of Responsible Party, 
     Government: Its Origins and Present State (1962) (providing 
     an excellent history of the American writers advocating 
     strong parties at the close of the 19th century and beginning 
     of the 20th century, including Woodrow Wilson, A. Lawrence 
     Lowell, Henry Jones Ford, and Frank Goodnow). Ranney also 
     offers an assessment of the widely held belief among 
     political scientists in the middle of the 20th century that 
     stronger political parties in the United States are 
     desirable, and discusses the conditions under which such 
     ``responsible'' parties can be achieved. See id. at 155-63.
     [FN7]. E.E. Schattschneider provides the classic description 
     of the struggle between parties and interest groups: The real 
     choice is between a strong party system on the one hand and a 
     system of politics in which congressmen are subjected to 
     minority pressures. The assumption made here is that party 
     government is better than government by irresponsible 
     organized minorities and special interests. The parties are 
     superior because they must consider the problems of 
     government broadly, they submit their fate to an election, 
     and are responsible to the public. E.E. Schattschneider, 
     Party Government 193 (1942). See also Morris P. Fiorina, The 
     Decline of Collective Responsibility in American Politics, 
     109 Daedalus 25, 27-28 (1980) [hereinafter Fiorina, Decline] 
     (providing a more contemporary formulation of 
     Schattschneider's argument); Anthony King, Running Scared 
     181-87 (1997) (discussing various ways to strengthen American 
     Political Parties).
     [FN8]. See Daniel Hays Lowenstein, On Campaign Finance 
     Reform: The Root of All Evil is Deeply Rooted, 18 Hofstra L. 
     Rev. 301, 348-51 (1989).
     [FN9]. American Pol. Sci. Ass'n, supra note 4.
     [FN10]. See id. at 9-11.
     [FN11]. See Norman Ornstein et al., Vital Statistics on 
     Congress, 1997-1998, at 11-13 (1999).
     [FN12]. See Nelson W. Polsby, Consequences of Party Reform 9-
     16 (1983); Austin Ranney, Curing the Mischiefs of Faction 82 
     (1975).
     [FN13]. See, e.g., Morris P. Fiorina, Whatever Happened to 
     the Median Voter? (May 11, 1999) (unpublished manuscript, on 
     file with the Columbia Law Review) (providing an excellent 
     summary of how different factors have contributed to 
     increased party discipline in the United States).
     [FN14]. See Table 1, infra.
     [FN15]. See Carroll J. Doherty, Campaign Finance Crusaders 
     Regroup After Latest Defeat, CQ Weekly Rep., Oct. 23, 1999, 
     at 2507-09.
     [FN16]. See id. at 2507.
     [FN17]. See Drew, supra note 2, at 15.
     [FN18]. See Michael J. Malbin & Thomas L. Gais, The Day After 
     Reform: Sobering Campaign Finance Lessons from the American 
     States 9-13 (1998).
     [FN19]. 424 U.S. 1 (1976).
     [FN20]. See Sorauf, supra note 2, at 9-11.
     [FN21]. See Drew, supra note 2, at 15; Frank J. Sorauf, Money 
     in American Elections 317-22 (1988).
     [FN22]. See Table 1, infra.
     [FN23]. See Martin P. Wattenberg, The Decline of American 
     Political Parties 1952-1988, at 58, 90-91 (1990).
     [FN24]. See Bruce Cain et al., The Personal Vote: 
     Constituency Service and Electoral Independence 9-12 (1989) 
     (providing a thorough assessment of this trend). For a 
     statistical measure of the rise of the personal vote, see 
     Stephen Ansolabehere et al., Old Voters, New Voters, and the 
     Personal Vote: Using Redistricting to Measure the Incumbency 
     Advantage, 44 Am. J. of Pol. Sci. 17 (2000).
     [FN25]. On the decline of party contact, see Steven 
     Rosenstone & Mark Hansen, Mobilization, Participation, and 
     Democracy in America 162-63 (1993).
     [FN26]. See Table 1, infra.
     [FN27]. See FEC, Reports on Financial Activity, various 
     years; Frank J. Sorauf, Money in American Elections 128-31 & 
     tbls. 5-1 & 5-2 (1988).
     [FN28]. See Herbert E. Alexander & Brian A. Haggerty, 
     Financing the 1984 Election 102-05 (1987).
     [FN29]. See Herbert E. Alexander & Monica Bauer. Financing 
     the 1988 Election 77-79 (1991).
     [FN301]. These figures are from two comprehensive studies of 
     all expenditures by all U.S. House and Senate candidates in 
     the 1990 and 1992 elections. See Sara Fritz & Dwight Morris, 
     Handbook of Campaign Spending: Money in the 1990 
     Congressional Races 5-6 (1991); Dwight Morris & Murielle E. 
     Gamache, Handbook of Campaign Spending: Money in the 1992 
     Congressional Races 6-7 (1994).
     [FN31]. See Drew, supra note 2. at 1-5: Allocation of Costs 
     for Voter Registration. [1976-1990 Transfer Binder] Fed. 
     Election Camp. Fin. Guide (CCH) P 5340 (FEC Advisory Opinion 
     1978-10).
     [FN32]. These figures combine the accounts of the DNC, the 
     RNC, the Democratic Senatorial Campaign Committee (``DSCC''). 
     the National Republican Senate Committee (``NRSC''), the 
     Democratic Congressional Campaign Committee (``DCCC''), and 
     National Republican Congressional Committee (``NRCC'').
     [FN33]. See FEC, Reports on Financial Activity, 1997-1998.
     [FN34]. See id.
     [FN35]. See FEC. Downloadable Databases Containing Financial 
     Information About Candidates, Parties and Other Committees 
     (visited Jan. 17, 2000) <http://www.fec.gov/finance/
newftp1.htm> (on file with the Columbia Law Review), FEC, 
     Reports on Financial Activity, 1993-1994, 1995-1996, and 
     1997-1998.
     [FN36]. Calculated by the authors based on Table 1, supra.
     [FN37]. Calculated from the second column of Table 1, supra.
     [FN38]. See Table 1. supra: Alexander & Bauer, supra note 29, 
     at 37.
     [FN39]. See FEC. Report on Financial Activity, 1997-1998.
     [FN40]. See id.
     [FN41]. See. e.g, Fiorina. Decline, supra note 7, at 3O-33; 
     Lowenstein, supra note 8, at 341-51.
     [FN42]. See. e.g, Morris Florina. Retrospective Voting in 
     American National Elections 202-03 (1981) (arguing in the 
     conclusion that without programmatic parties, retrospective 
     voting is not possible).
     [FN43]. See Ornstein et al., supra note 11, at 61-62.
     [FN44]. David Mayhew. Congressional Elections: The Case of 
     the Vanishing Marginals, 6 Polity 295, 295(1974).
     [FN45]. See Ansolabehere et al., supra note 24, at 30; Steven 
     D. Levitt & Catherine D. Wolfram, Decomposing the Sources of 
     Incumbency Advantage in the U.S. House, 22 Legis. Stud. Q. 
     45, 46 (1997).
     [FN46]. See Gary C. Jacobson, Money in Congressional 
     Elections 105-23 (1980); Stephen Ansolabehere & James M. 
     Snyder, Jr., Money and Office, in Continuity and Change in 
     Congressional Elections (David Brady et al. eds., forthcoming 
     2000).
     [FN47]. Of course, it is not clear why increasing 
     competitiveness, per se, is a good thing--rather, it depends 
     on why there is little competition. If it is due to entry 
     barriers of some sort, then it is probably bad. If it is due 
     to the effectiveness of incumbents in giving constituents 
     what they want, then it may not be.
     [FN48] This pattern is noted by Gary C. Jacobson, Party 
     Organizations and the Distribution of Campaign Resources, 100 
     Pol. Sci. Q. 603, 604 (1985-86), and by Paul S. Herrnson, 
     National Party Decision Making, Strategies, and Resource 
     Distribution in Congressional Election, 42 W. Pol. Q. 301, 
     307-09 (1989).
     [FN49]. See Figure 1, infra.
     [FN50]. See Ornstein et al., supra note 11, at 61-62.
     [FN51]. See Stephen Ansolabehere & James M. Snyder, Jr., 
     Money and Institutional Power. 77 Tex. L. Rev. 1673. 1682-98 
     (1999); James M. Snyder. Jr., The Market for Campaign 
     Contributions: Evidence for the U.S. Senate 1980-1986, 5 
     Econ. & Pol. 219, 219 (1994).
     [FN52]. See Ansolabehere & Snyder, supra note 46.
     [FN53]. See FEC, Report on Financial Activity, various years.
     [FN54]. See Stephen Ansolabehere & James M. Snyder, Jr., 
     Money, Elections, and Candidate Quality (1997) (unpublished 
     manuscript) (on file with Columbia Law Review). Alan Gerber, 
     Estimating the Effect of Campaign Spending on Senate Election 
     Outcomes Using Instrumental Variables, 92 Am. Pol. Sci. R. 
     401 (1998), produces very similar estimates for the U.S. 
     Senate.
     [FN55]. See Lowenstein, supra note 8, at 363-64.
     [FN56]. See Kevin M. Leyden & Steven A. Borrelli, Party 
     Contributions and Party Unity: Can Loyalty Be Bought? 43 W. 
     Pol. Q. 343, 351-52 (1990) (concluding on the basis of Tobit 
     results that ``the Democratic Party, `rewards' its most loyal 
     members with greater financial aid'').
     [FN57]. See, e.g., David M. Cantor & Paul Herrnson, Party 
     Campaign Activity and Party Unity in the U.S. House of 
     Representatives, 22 Legis. Stud. Q. 393, 402 (1997) 
     (concluding from the results of regression analyses that 
     ``past party unity has no significant effect on the 
     distribution of party assistance in campaign fundraising. . . 
     .'').
     [FN58]. See Kevin M. Leyden & Steven A. Borrelli, An 
     Investment in Goodwill: Party Contributions and Party Unity 
     Among U.S. House Members in the 1980s, 22 Am. Pol. Q. 421, 
     421-52 (1994) (reporting an association between party 
     contributions to candidates and the likelihood that the 
     legislator voted similarly to the party leadership). The 
     problem of baselines is that the research does not 
     simultaneously establish how the politician would have voted 
     in the absence of contributions. If, for example, a 
     legislator represented a district similar to the district of 
     a party leader, then that legislator might cast roll call 
     votes in line with the leader not because of party 
     contributions but because such votes represent his or her 
     constituents.
     [FN59]. Tobit regression predicts a dependent variable, in 
     this case contributions to and expenditures on behalf of 
     individual candidates that came from political party 
     committees, as a linear combination of many independent 
     variables predicting a single dependent variable. Tobit 
     regressions correct for ``censoring'' effects due to many 0's 
     in the dependent variable. See generally William H. Greene, 
     Econometric Analysis 962-65 (3d ed. 1997). The regressions 
     include year dummies, district partisanship. opponent total 
     expenditures. and roll call voting scores.
     [FN60]. Party money in this analysis consists of 
     contributions and coordinated expenditures combined. The 
     average is for incumbents who received at least some party 
     money contributions. The average for incumbents and non-
     incumbents combined was $14,000.
     [FN61]. In the regressions pooling all years, for Democrats, 
     the coefficient for proximity of roll call voting to the 
     party median is positive and highly significant, with a t-
     statistic over 6. Looking at each year separately, the effect 
     of proximity to the median for Democrats is significant in 
     all years. For Republicans, the effects in the pooled data 
     and in each year are always insignificant, and in some years 
     have the ``wrong'' sign.
     [FN62]. FEC, Press Release, FEC Reports on Political Party 
     Activity for 1997-1998, at 2, Tables 1-2 (Apr. 9, 1999) 
     http://www.fec.gov/press/ptyye98.htm (on file with the 
     Columbia Law Review).
     [FN63]. See id.; FEC, Press Release, FEC Reports on Political 
     Party Activity for 1995-1996, at Tables 3-4 (Mar. 19, 1997) 
     http://www.fec.gov/1997news.html (on file with the Columbia 
     Law Review).
     [FN64]. See Idaho Secretary of State, Election Division: 
     Campaign Finance (visited Jan. 17, 2000) <http://
www.idsos.state.id.us/elect/finance.htm> (on file with the 
     Columbia Law Review) [hereinafter Idaho Election Div. 
     Website]; North Carolina State Board of Elections, Campaign 
     Finance Reports Download (visited Jan. 17, 2000) <http://
www.sboe.state.nc.us/cro/finance.htm> (on file with

[[Page E269]]

     the Columbia Law Review) [hereinafter N.C. Bd. of Elections 
     Website]. In Ohio, data are distributed by the Secretary of 
     State, 30 East Broad Street, 14th Floor, Columbus, Ohio 
     43266-0418. A database was provided by the Secretary of 
     State's office.
     [FN65]. See Michael Barone & Grant Ujifusa, The Almanac of 
     American Politics, 2000 1240-48 (1999) (statistics on page 
     1248).
     [FN66]. See id. at 1184-93 (statistics on page 1193).
     [FN67]. See id. at 506-12 (statistics on page 512).
     [FN68]. See FEC, Campaign Finance Law 98, Chart 1 (1999).
     [FN69]. See Idaho Election Div. Website, supra note 64 (for 
     hardcopy versions of these data, see State of Idaho, 
     Contributions and Expenditures of Candidates for Statewide, 
     Legislative, and Judicial Office and Political Committees 
     (Jan. 1, 1997-Dec. 31, 1998)).
     [FN70]. See North Carolina State Board of Elections, Analysis 
     of Contributions and Expenditures, 1998 (Jan. 15, 1999).
     [FN71]. See Ohio Secretary of State, Ohio 1998, Campaign 
     Finance Facts 65-69.
     [FN72]. See Barone & Ujifusa, supra note 65 at 512, 1184, 
     1248 (reporting population figures). Transfers to state 
     committees compiled by authors from FEC, Press Release, FEC 
     Reports on Political Party Activity 1997-1998, supra note 61, 
     at Tables 9 and 10.
     [FN73]. These figures come from FEC, Press Release, FEC 
     Reports on Political Party Activity 1997-1998, supra note 61, 
     at Tables 9 and 10.
     [FN74]. Calculated by the authors from reports of the state 
     election commissions. See supra notes 56-56.
     [FN75]. See Ohio Secretary of State, supra note 71. We were 
     able to classify 94 percent of the 41,000 expenditure items 
     listed in the database into categories of office and 
     overhead, fundraising, grassroots activities, media, 
     contributions, and payroll.
     [FN76]. See id.
     [FN77]. See Fritz & Morris, supra note 30, at 15-29; Morris & 
     Gamache, supra note 30, at Chapter 2.
     [FN78] These estimates are based on data sets provided by the 
     relevant state offices.
     [FN79]. These figures are calculated from the datasets 
     provided by the relevant state offices. See supra note 50.
     [FN80]. The 20 cents figure is an average of the reduction in 
     expenditures in the three states combined.
     [FN81]. FEC, Press Release, Reports on Political Party 
     Activity for 1997-1998, supra note 61, at 2.
     [FN82]. FEC, Federal Elections 96: Election Results for the 
     US President, the US Senate, and the US House of 
     Representatives, at tbl. 1 (1997).
     [FN83]. See Alan Gerber, A Tale of Two Literatures (Mar. 
     1999) (unpublished manuscript, on file with Dept. of 
     Political Science, Yale University); Alan Gerber & Donald 
     Green, The Effect of a Nonpartisan Get-Out-the-Vote Drive: An 
     Experimental Study (Aug. 1999) (unpublished manuscript, on 
     file with Dept. of Political Science, Yale University) 
     [hereinafter Nonpartisan GOTV]; Alan Gerber & Donald Green, 
     Does Canvassing Increase Voter Turnout? A Field Experiment, 
     96 Proc. Nat'l Acad. Sci. 10939 (1999).
     [FN84]. See Gerber & Green, Nonpartisan GOTV, supra note 83, 
     at 220.
     [FN85]. We are indebted to Alan Gerber for assisting with 
     these calculations.
     [FN86]. See, e.g., Lowenstein, supra note 8, at 351-55; 
     Fiorina, Decline, supra note 7, at 26-27.
     [FN87]. See Ansolabehere & Snyder, supra note 46.

                                  ____
                                  

 [Institute of Governmental Studies and Citizens Research Foundation--
                       Working Paper, July 2000]

     Soft Money Spending by State Parties: Where Does It Really Go?

                   (By Ray La Raja and Karen Pogoda)


                                Summary

       In this study we analyze campaign expenditures by state 
     political parties from the 1992 through 1998 elections, which 
     includes disbursements of soft and hard money. We find 
     evidence to support a more complex reality about how soft 
     money is used by parties than is typically conveyed in the 
     news media. While party spending on issue ads increased 
     dramatically in 1996 and 1998, so did party-building 
     activities, such as voter mobilization and grassroots, which 
     were encouraged by amendments to the Federal Election 
     Campaign Act in 1979. We also find that Democratic state 
     parties spend more soft money than Republican parties on 
     media-related activities, such as issue ads, probably to 
     compensate for their lack of hard relative to the 
     Republicans. We conclude with a recommendation that reformers 
     consider some of the positive effects on American elections 
     of party control of campaign resources as they attempt to 
     curb the potential for corruption by restricting or 
     eliminating soft money contributions to parties.


                              Introduction

       Scarcely a week passes during an election year without news 
     reports of a corporation or wealthy individual making a large 
     soft money contribution to one of the major parties. Election 
     web sites sponsored by nonpartisan organizations and 
     government agencies routinely provide access to data on 
     campaign contributions to candidates and parties. This 
     widespread focus on contributions to and from political 
     committees stems from a genuine concern to expose corruption 
     rooted in the exchange of money. Even without sufficient 
     evidence of corruption, reform advocates continue a single-
     minded quest to restrict the size of political contributions, 
     without looking at the other side of the equation. What do 
     candidates and their parties do with campaign contributions? 
     Are they spent in ways that encourage or dampen competition? 
     Does party soft money spending generate any public benefits 
     in elections, beyond its intended support for candidates?
       A narrow focus on the sources of contributions prevents us 
     from speaking to such questions. In this paper, we try to 
     redress what we see as a one-sided approach to the study of 
     campaign finance, particularly with respect to the soft money 
     issue. We set out to answer a simple question: how do 
     political parties spend soft money? By most journalistic 
     accounts, the conclusion is that parties use soft money to 
     pay for ``issue ads'' that support the presidential or 
     congressional candidates. Our study demonstrates this partial 
     truth, but also provides evidence to support a more complex 
     reality. In fact, the parties continue to spend a great deal 
     of soft money on traditional party-building functions that 
     mobilize voters through individual contacts.
       Why should we care about making such distinctions about 
     party spending? When Congress amended the Federal Election 
     Campaign Act (FECA) in 1979, it made provisions for parties 
     to spend unlimited amounts on so-called party building 
     functions. The earlier version of the FECA in 1974 inhibited 
     state and local parties from participating in the 
     presidential campaign through grassroots activities because 
     of rules limiting contributions to the candidates. The 1979 
     amendment, which exempted generic party activity from 
     contribution limits for the presidential campaign, was a 
     deliberate effort to increase the party role in American 
     elections. In this study we find that this policy, worked. 
     State parties, in fact, increased mobilization and grassroots 
     activities in the 1990s, largely as a result of the 1979 
     exemption and the increased use of soft money.
       It is unlikely, however, that when Congress made changes to 
     the FECA, members understood the role that soft money would 
     play in paying for issue advocacy, the generic media 
     advertising sponsored by parties that often crosses the line 
     into direct candidate support. Reform advocates argue, with 
     merit, that issue advocacy reduces the distinction between 
     hard and soft money spending. By producing campaign ads that 
     bolster a particular candidate in all but name, parties found 
     a way to get around limits on candidate support. So long as 
     the party avoids the electioneering phrase, ``vote for,'' or 
     something similar, they can pay for these ads with soft 
     money. If parties
       We find conclusively that national parties exploited an 
     opportunity to help their nominees for federal offices by 
     channeling funds to state parties for the express purpose of 
     purchasing issue ads. Party-sponsored issue ads increased 
     dramatically in the 1996 and 1998 elections, just when 
     national parties were transferring significant sums of soft 
     money to state parties. We also demonstrate that most media-
     related spending occurred in states with competitive races 
     for the 1996 presidential and 1998 Senate campaigns.
       But our analysis also reveals that party issue ads are only 
     one part of the story. While expenditures on media-related 
     activities surged in 1996, so did spending on grassroots and 
     voter mobilization efforts--the kind of party campaign 
     activity Congress wanted to encourage when it revised the 
     FECA in 1979.
       We believe our findings complicate the reform debate 
     considerably. On the one hand, we observe parties violating 
     the spirit, if not the letter, of the law when they pay for 
     issue ads with soft money that help federal candidates. And 
     yet, we also notice that soft money has been used to bolster 
     party activities that citizens, elected officials and 
     political scientists view as positive for democracy. The 
     increased use of soft money is associated with greater 
     spending on political rallies, bumper stickers and yard 
     signs, as well as voter identification and get-out-the-vote 
     programs.
       Another healthy sign, especially from the perspective of 
     political scientists, is that state party organizations 
     appear to be growing stronger, if somewhat more reliant on 
     national organizations. Our findings demonstrate that 
     infusions of soft money have augmented activities at party 
     headquarters, as evidenced by increased spending on staff 
     salaries, rent, computers, telephones and other 
     organizational maintenance necessities. For several 
     generations, scholars have worried about the demise of party 
     organizations that formerly played a key role in nominating 
     candidates and pulling together coalitions. Weak parties 
     leave the field open to single-issue interest groups and 
     candidate-centered campaigns that tend to fragment the 
     electorate and subsequently increase the difficulty of 
     governing. To the extent that party organizations are 
     increasingly active in campaigns, we take this as a positive 
     sign of party revitalization. Beyond our preliminary 
     analysis, future research should investigate in greater 
     detail the degree to which party activity reflects ``pass 
     throughs'' of money for specific candidates or support for a 
     collective and unifying form of campaigning, closer to the 
     model of responsible parties outlined by the American 
     Political Science Association.
       We make no assertions about whether soft money strengthens 
     the party system and improves the electoral process. Our 
     findings are merely suggestive. In part, we publish the 
     results of this working paper to give pause to supporters of 
     a ban on soft money from the campaign finance system. By 
     moving too quickly to eliminate party resources, the public 
     may forego potential benefits of stronger parties. Worse, the 
     money that now flows through parties may simply be re-
     channeled through other, less visible organizations. 
     Experience shows this is not simply plausible but probable. 
     The prospects for effective reform are enhanced through a 
     genuine understanding of the outputs, as well as the inputs, 
     of campaign money.


                            Some Background

                          What is soft money?

       Soft money is a term developed in the 1980s to 
     differentiate contributions to the party

[[Page E270]]

     that may be used to support federal candidates directly from 
     those that cannot. Under federal law, the purpose of soft 
     money is for party building and not for direct candidate 
     support. In 1974, when Congress passed amendments to the 
     Federal Election Campaign Act (FECA), it imposed a limit on 
     contributions to the party, and the amount of direct support 
     that parties could provide their candidates, either through 
     cash or in-kind contributions. Individuals could donate no 
     more than $20,000 to parties, and PACs were limited to 
     $15,000. In the late 1970s, leaders of state party 
     organizations lobbied Congress and the Federal Elections 
     Commission (FEC) to permit them to extend the use of soft 
     money to generic party activities that included distribution 
     of lawn signs, bumper stickers and activities aimed broadly 
     at mobilizing the vote. They argued that federal laws 
     limiting party support of presidential candidates constrained 
     them from performing generic party campaign activities that 
     broadly benefited both federal and state candidates. Congress 
     responded with amendments in 1979 permitting state and local 
     parties to spend unlimited funds on ``party-building'' 
     activities, such as grassroots campaign materials and voter 
     contact activities. It is important to note that Congress did 
     not authorize state committees to use unregulated funds to 
     pay for these activities. State parties were required to use 
     funds raised under the rules of the FECA.
       In fact, the so-called soft money loophole did not open 
     until the FEC was faced with the dilemma of providing 
     accounting guidelines to state parties where state laws 
     permitted unrestricted contributions from unions and 
     corporations. In response to a query by the Republican State 
     Committee of Kansas about how to allocate federal and 
     nonfederal expenses incurred by party building activities, 
     the FEC declared that the Kansas Republicans could use their 
     nonfederal fund to pay a reasonable estimate of the 
     nonfederal share of cost. This ruling effectively meant that 
     the party could use a nonfederal fund--which had no 
     constraints on corporate or union contributions under Kansas 
     law--to fund activities that benefited, in part, federal 
     candidates. A 1988 U.S. District court order, pursued by 
     reform activists at Common Cause, required the FEC to provide 
     detailed allocation requirements to prevent the parties from 
     abusing their new ability to use soft money in federal 
     elections. Yet even with the promulgation of specific 
     allocation requirements, the national and state parties 
     continued to seek the advantages of permissive state campaign 
     finance laws to raise and spend nonfederal funds to support 
     their federal candidates through party-building activities.
       Since raising unregulated soft money is easier than federal 
     (hard) money, which has contribution limits, the national 
     parties pushed to expand the definition of party-building
       Although the FEC attempted to curtail the use of issue ads 
     and other party activities that crossed the line from party 
     building to candidate support, they were blocked by a Supreme 
     Court ruling, Buckley v. Valeo. In this case, the justices 
     tried to distinguish between constitutionally-protected free 
     speech and electioneering messages. The ruling demonstrated 
     that the courts would narrowly define ``electioneering'' to 
     include messages that clearly exhorted citizens to vote for 
     or against specific candidates. Under a narrow definition, 
     parties could safely use soft money for issue ads that helped 
     candidates so long as they avoided electioneering language 
     that constituted ``express advocacy'' for a candidate. Such 
     language includes use of the words, ``vote for,'' ``oppose,'' 
     ``support,'' and the like.
       One consequence of Buckley was a deliberate party strategy 
     to funnel money to state parties where complex rules 
     permitted them greater use of soft money. In a presidential 
     election year, national committees must allocate hard money 
     to at least 65 percent of administrative costs. The state 
     parties, in contrast, might pay for the same activity with as 
     little as 25 percent hard money, depending on a formula that 
     considers the ratio of state and federal candidates in the 
     election. Much has been written about party efforts to 
     conceal campaign advertising behind the shroud of state party 
     building, but there has been little systematic analysis to 
     demonstrate the extent of this activity. We collected 
     financial data on the 100 state parties over four elections 
     to examine how parties use soft money.


                                Methods

       Our analysis is based on expenditure data provided by the 
     Federal Elections Commission (FEC). Since the 1992 election 
     cycle, parties at all levels have been required to maintain 
     two separate accounts, federal and non-federal. The non-
     federal account is not reported to the FEC because these 
     funds are applied exclusively to nonfederal activities, such 
     as party support for state legislative candidates. The 
     federal account, however, must include itemized expenditures 
     that potentially benefit a federal candidate, even if the 
     spending also helps state and local candidates as well. The 
     FEC calls this ``Joint'' spending. Party treasurers are 
     required to allocate hard and soft money for joint spending 
     to reflect the federal-nonfederal split of benefits to 
     candidates. To limit the discretion of treasurers--who have 
     an incentive to claim that benefits accrue mostly to state 
     and local candidates so as to avoid using hard money--the FEC 
     promulgated rules determining the proper mix of hard and soft 
     funds for a given kind of joint activity. For example, 
     administrative costs are allocated according to the ratio of 
     federal candidates to total candidates (state and federal) in 
     the state. We use the federal account data, with its matching 
     hard and soft allocations, to determine how parties spend 
     soft money.
       We believe the federal account provides us with the greater 
     part of party expenditures. The non-federal account, 
     according to some estimates, accounts for at least an 
     additional 25 percent in soft money that state parties spend 
     exclusively to benefit state and local candidates. State 
     parties are compelled by federal law to use federal accounts 
     whenever they perform some kind of generic party activity 
     that might jointly benefit party candidates up and down the 
     ticket. This requirement ensures that every expense, from 
     routine office costs to voter identification programs, shows 
     up in the federal account. The federal account also includes 
     itemized expenditures on media that parties call ``issue 
     advocacy.'' It is precisely because parties claim that issue 
     advocacy reflects party rather than candidate specific themes 
     that they must report this activity as generic (or joint) in 
     the federal account.
       Our study looks at the federal reports submitted to the FEC 
     by the 100 state parties, for election cycles 1992 through 
     1998. Fortunately, staff at the FEC entered, by hand, each 
     expenditure item in database files from the hard-copy reports 
     submitted by state parties. Using these files, we developed a 
     coding scheme to categorize more than 300,000 itemized 
     expenditure entries in each election cycle. The categories 
     are the following:
       Overhead: office related expenses such as rent, salaries, 
     computers, travel, and utilities.
       Media: communication expenditures for television, radio and 
     newspaper and production and purchase costs.
       Mobilization: costs of contacting individual voters through 
     direct mail, telephone banks, canvassing and voter 
     identification files.
       Grassroots: activities that encourage citizen participation 
     in campaigns. Expenditures for rallies, fairs, volunteer 
     precinct walks, banners, slate cards, bumper stickers, and 
     local party support.
       Multi-candidate contributions: non-generic in-kind 
     contributions from the party to several candidates, e.g., 
     newspaper ads, that jointly benefit specific federal and 
     state candidates. These are distinct from the direct 
     contributions to candidate committees.
       Fundraising: costs associated with joint fundraising for 
     federal, state and local campaigns.
       Unidentified: expenditures that could not be determined 
     from FEC reports.
       In the following sections, we provide summaries for total 
     soft money expenditures in each of the above categories. We 
     are able to compare the data over four election cycles, 1992 
     through 1998.


                                Findings

            Are the state parties spending more soft money?

       There is little doubt that state parties are more active 
     than ever in election campaigns. Combined soft and hard money 
     spending in the state party federal accounts almost doubled 
     between 1992 and 1996. Undoubtedly, some of this spending is 
     the product of mere ``pass throughs,'' the transfers from the 
     national to state parties to purchase issue ads and other 
     services in support of federal candidates. But as we 
     demonstrate later, state parties have also increased spending 
     on campaign activities that serve party building functions.
       Much of this growth in spending has been spurred by 
     additional use of soft money. In the 1996 presidential 
     election the 100 state parties spent $178 million, almost 
     triple the amount of soft money, spent in 1992. Similarly, 
     between the 1994 and 1998 midterm elections the parties 
     doubled their use of soft money, spending a record $187 
     million. Hard money expenditures have also risen but not at 
     the same rate. Since FEC rules require soft-hard matching for 
     each campaign activity, it is not surprising that hard money 
     spending increases with soft money spending. It appears, 
     however, that soft money pays for a larger portion of 
     activities with each passing election cycle. In 1998, for the 
     first time since 1992 when state parties were required to 
     report soft money spending, they spent more soft than hard 
     money in their federal accounts.
       The apparent shift from hard to soft money is not difficult 
     to explain. Soft money is easier to obtain since there are no 
     limits on contributions to parties, except when state laws 
     regulate party fundralsing. A party that wants to preserve 
     its hard money for candidate contributions and coordinated 
     expenditures in federal elections will purchase goods and 
     services with soft money whenever possible. Over the four 
     most recent election cycles, the state parties have learned 
     how to match soft and hard money

[[Page E271]]

     expenditures to maximize the use of the former. One 
     indication that parties behave this way is that direct state 
     party support for federal candidates, mostly in the form of 
     coordinated expenditures increased from $5 million in 1996 to 
     $18 million in 1998. We suspect state parties substituted 
     soft for hard money when paying for many kinds of campaign 
     activities, thereby freeing up additional hard money for 
     direct candidate support.
       An important question to ask is whether soft money reported 
     in the federal accounts of state parties is actually 
     controlled by the national parties, whose primary interest is 
     to elect candidates for federal office. To the extent that 
     national party supports the state parties through transfers, 
     we can make the inference that they have some control over 
     state party expenditures. Table 1 (not supplied) gives a 
     sense of how much state parties rely financially on the 
     national parties. The national parties support a larger 
     percentage of state party budgets in 1996 and 1998 than they 
     did earlier, suggesting that they have more influence in 
     state party affairs than in earlier elections. Prior to 1996, 
     national party transfers did not account for more than 14% of 
     the federal accounts of state committees. In the 1996 and 
     1998 elections, this portion grew to 42% and 31% 
     respectively. Table 1 also illustrates that state parties 
     rely more heavily on national parties for soft money than 
     hard money. National parties provide just under a quarter of 
     the hard money that state parties end up spending, but 65% of 
     the soft money they spent in 1996 and 37% in 1998. It appears 
     that soft money has become a primary means of intra-party 
     support. State parties continue to raise the majority of 
     funds on their own--indeed, they raise more money 
     independently than ever before--but they receive significant 
     support from the national parties. In addition to party 
     transfers, some journalistic accounts report that state 
     parties benefit from soft money contributors who are 
     encouraged to donate to state parties by officials of the 
     national party.
       Since national parties provide as much as one-third of 
     state party funds, we suspect that portions of soft money 
     from the national parties are targeted to achieve national 
     party goals, which may differ from the priorities of state 
     organizations. These data demonstrate unequivocally that the 
     direction of resource flows between parties has reversed 
     since the 1960s, when national parties had to solicit 
     contributions from state affiliates. Heard (1960) predicted 
     such a change would create opportunities for party 
     integration and growth, even as it augmented tensions among 
     levels of party.
       To summarize, soft money spending by state parties has 
     risen each year since 1992, and outpaced hard money spending 
     in 1998. FEC matching requirements will ensure that soft 
     money spending does not entirely eclipse hard money spending, 
     but it appears parties exploit allocation rules to spend soft 
     rather than hard money. We should note, however, that state 
     parties raise and spend increasing sums of hard money, funds 
     that meet all the requirements of the FECA. Hard money 
     spending doubled between the 1992 and 1996 elections and the 
     state parties are responsible for raising three-fourths of 
     this money themselves. The prospect of securing soft money 
     from the national parties may spur state parties to engage 
     more effectively in raising hard money, precisely because of 
     the federal matching requirements. We also find preliminary 
     evidence that

            How do state political parties spend soft money?

       We now turn to a description of how state parties use soft 
     money in campaigns. As we stated earlier, there is anecdotal 
     evidence, mostly from the news media, describing the use of 
     soft money for issue ads. More systematic scholarly research 
     demonstrates that in key races soft money is invested in the 
     ``ground war'' of campaigns, through contacts with individual 
     voters using direct mail and telephone banks. Party and 
     campaign finance scholars continue to speculate whether the 
     infusion of soft money in the last two decades has altered 
     patterns of state party activity. Advocates of stronger 
     parties have argued that providing parties with privileged 
     access to campaign resources would reverse the long decline 
     of party organizations. From their perspective, the 
     introduction of soft money into the party system provides an 
     interesting test case for this theory. How will parties 
     behave with this new wealth generated by soft money? Will 
     they spend additional increments to build the party through 
     voter identification programs and grassroots activity? Or 
     will soft money simply buttress candidate-centered campaigns, 
     with the parties serving as pass-throughs to pay for 
     television ads promoting individual nominees?
       Our findings will hardly satisfy those who seek support for 
     an opinion that soft money is either good or bad for the 
     party system. In fact, we find elements of what some would 
     consider ``bad'' as well as ``good'' spending. On the 
     positive side, we observe that state organizations continue 
     to use funds in ways we traditionally expect of parties: to 
     mobilize voters, provide grassroots paraphernalia like bumper 
     stickers and lawn signs, and, of course, for basic 
     organizational maintenance activities such as paying rent and 
     salaries (overhead) and fundraising. In short, soft money 
     enables parties to spend additional resources on party-
     building activities.
       The election in 1996, however, marked a dramatic shift 
     toward greater spending on media related activities. Whereas 
     the state parties spent just 3 percent of their budgets on 
     media activities in the 1992 presidential election year, four 
     years later this category absorbed more than one-third of 
     their budgets. The shift is more striking in absolute terms: 
     media spending jumped from about $2 million to $65 million. 
     The reasons for this shift have been explained in many 
     journalistic accounts of the 1996 and 1998 campaigns. The 
     increase in media spending in 1996 was a result of campaign 
     strategies pursued by the parties and presidential candidates 
     to saturate critical electoral markets with televised issue 
     ads that benefited the candidates in all but name. Dick 
     Morris, the key Clinton-Gore campaign strategist, urged the 
     DNC to begin televising issue ads in the summer and early 
     fall as a way to shore up a faltering Clinton early in the 
     election and undercut the presumptive GOP nominee, Bob Dole. 
     The RNC, in support of the Dole-Kemp ticket, countered with 
     the same strategy right before and after the convention in 
     July. Apparently, both national parties tried to take 
     advantage of the favorable soft-hard ratios available to 
     state parties by delegating responsibility for purchasing the 
     ads to the latter.
       Ironically, soft money spending on issue ads might be an 
     artifact of the sweeping reforms of 1974 that established a 
     system of public financing for presidential candidates. If a 
     candidate accepts public funding, in the primary he faces 
     limits on spending in each state. A competitive race could 
     cause candidates to bump up against these limits rather early 
     in the primary season, especially given the trend toward 
     front-loading of primaries, forcing them to curtail spending 
     severely during the weeks leading up to the convention. Bob 
     Dole, for example, faced several tough and well-funded 
     challengers in 1996. He was forced to spend money fending off 
     Gramm, Buchanan and Forbes. Clinton, in contrast, began using 
     party soft money, as well as primary campaign funds, to 
     attack the GOP and promote his campaign themes for the 
     general election. Dole and the Republicans could only 
     retaliate with party soft money ads, given that the candidate 
     would not receive additional public funds for the general 
     election until after the convention. The late timing of FEC-
     released public funds leaves a good part of the summer in 
     which either candidate can harm the other through attack ads. 
     The parties joined in the campaign, in part, to bridge the 
     period between the point at which a nominee effectively, but 
     not officially, wins the party nomination, and the official 
     start of the general election season as determined by the end 
     of the party conventions.
       The increasing use of soft money for issue ads may also 
     reflect the inadequacy of a public funding system for 
     presidential campaigns that falls to keep pace with rising 
     media costs. A standard thirty-second advertisement during 
     prime time in a major media market can cost in the range of 
     $20,000 to $30,000. Only fifteen years ago, the same ad cost 
     approximately half that amount. Although presidential funding 
     system adjusts for inflation, average media unit costs have 
     risen faster than the average for all other goods and 
     services. More importantly, according to one study, campaign 
     strategists rely increasingly on expensive media-related 
     activities, especially television, which drive up the cost of 
     the entire campaign.
       During midterm elections, spending on media decreases 
     without the demands of a national campaign. In the 1998 
     midterm, the amount spent on media related activities by 
     state parties was cut more than half, to $30 million from two 
     years earlier. But this amount was ten times as much as party 
     spending on similar activities in the 1994 midterm election. 
     The lessons of using party soft money for issue ads in the 
     1996 presidential campaign had obviously been passed on for 
     congressional elections. According to a study sponsored by 
     the Brennan Center, party spending on issue ads--which 
     includes both state and national organizations--amounted to 
     $25.9 million. This spending accounted for close to 45,000 
     ads, reflecting about 20 percent of all campaign advertising.
       Our data demonstrate clearly that soft money was 
     transferred to state parties to fund media-related activity 
     that comprised mostly issue ads. But assuming that every 
     dollar transfer produced a dollar's worth of issue ads, the 
     fact remains that state parties spent little more than 55 
     percent of transfers on issue ads in 1996, and 43 percent on 
     them in 1998. Where did the rest of the soft money go? The 
     answer is that parties used ``excess'' soft money to increase 
     traditional party activities. In 1996, spending on voter 
     mobilization almost doubled from the previous presidential 
     election, rising from $8 million to $16 million. Over the 
     same period, spending on grassroots activities increased 
     sevenfold, from $1.2 to $8.3 million.
       These figures, of course, are small in comparison to 
     allocations for media-related activity. One reason is that 
     the cost of bumper stickers, or even telephone banks, is 
     considerably less than that of media-purchases in 
     metropolitan markets. At about ten cents per bumper sticker, 
     one million dollars will purchase 10 million bumper stickers. 
     The same amount will provide about forty ads (30 seconds) on 
     network TV in a major media market during prime time.
       Importantly, media spending did not crowd out spending on 
     traditional party activities. The portions of the party 
     budget spent on mobilization and grassroots did not change 
     substantially even when media spending soared. In the 1998 
     elections, Magleby (2000)

[[Page E272]]

     reports that the parties, particularly the Democrats, 
     emphasized a ``ground war'' strategy that involved lots of 
     direct mail, telephone banks and other get-out-the-vote 
     activities. It appears, according to Table 2, that parties 
     used additional soft money in 1998 to intensify mobilization 
     efforts, spending nearly the same portion of their budget on 
     such activities as they did in 1992 and 1994.
       Additional soft money has also been used to expand party 
     headquarter operations. In 1992, state parties spent $42 
     million on overhead, which include payments for salaries, 
     rent and other organizational maintenance costs. By 1998, 
     this total had risen to $107 million. Certainly, we would 
     want to know the degree to which these rising expenses at 
     headquarters reflect sustained organizational growth or 
     temporary surges in activity for the limited campaign season. 
     An analysis of party budgets during the off-election year 
     should resolve whether these costs reflect enduring 
     investments in the party organization. At the very least, the 
     rising costs associated with maintaining party headquarters 
     suggests that state party organizations are a locus of 
     increased campaign activity.

                         Partisan differences?

       To see if parties pursue different strategies with soft 
     money we compare them for the 1996 and 1998 elections. The 
     parties appear to spend similar amounts on all activities 
     except for media, which accounts for much of the Democratic 
     lead. In 1996 the Democratic state parties allocated about 
     $48 million for media, three times as much as the 
     Republicans. The gap for the 1998 midterm election was not as 
     great since neither party spent as much on media, but the 
     Democratic state parties continued to outspend the 
     Republicans at the state level by more than 6 million. We 
     believe these partisan differences exist because the national 
     Democrats, being the relatively poorer party, attempt to 
     exploit soft money for federal races more than Republicans. 
     They do this by transferring soft money to state parties 
     where the spending ratios for soft and hard money are higher, 
     meaning that the state parties can use more soft money than 
     the national parties to pay for the same activity.
       The practice of using the state parties for national party 
     goals probably comes at a cost. State parties might dun the 
     national parties for these services by requesting additional 
     transfers of soft money beyond the costs of the services. At 
     the very least, a transfer strategy imposes greater 
     coordination costs on national parties, particularly the 
     Democrats, who appear to do this more often. National parties 
     must monitor the transferred funds to ensure state parties 
     spend them properly. The national Republicans, with a 
     significant advantage in hard money receipts, can more likely 
     avoid this problem by producing and purchasing media services 
     directly, even if they must pay with additional hard money. 
     We suspect that the national committees of the Republican 
     Party outspend their Democratic counterparts on such campaign 
     activities.
       The Democratic strategy of transferring soft money to state 
     parties for issue ads is clearly evident from Table 4, not 
     supplied, which lists states with the highest media-related 
     spending. In each of these states there was a close federal 
     electoral contest. In 1996, Ohio was not


                               Conclusion

       We began with a question about how parties spend soft 
     money. We speculated that soft money was not simply a 
     resource to fund issue ads, but also a primary means to 
     support party organizations and their traditional campaign 
     activities. Our finding is that parties use soft money in 
     ways that would strike many observers--including those 
     favoring a ban on soft money--as positive. This preliminary 
     study illustrates that parties use soft money to invest in 
     campaign activities that promote party-building and citizen 
     participation. If soft money permits the party to reach 
     additional voters through telephone calls and mail, or 
     generate enthusiasm for political campaigns through rallies 
     and yard signs, then perhaps we are shortchanging American 
     campaigns by cutting off this supply of money. The 
     overemphasis in the news and by public interest advocates on 
     the media strategies of parties obscures the fact that 
     parties do many things with soft money.
       Undoubtedly, parties also exploit soft money to fund issue 
     ads through their state organizations. Media-related spending 
     by state parties jumped from just $2 million in 1992 to $65 
     million in 1996. The Democrats appear to take advantage of a 
     state-sponsored issue ad strategy more than the Republicans, 
     probably because they trail the Republicans in raising hard 
     money. Both parties, however, use most of their soft money to 
     expand party headquarter operations during the campaign. 
     Since 1992, they have more than doubled the amount spent 
     contacting individual voters through various voter 
     identification and get-out-the-vote programs. In the last 
     midterm election, just 16% of soft money went toward issue 
     ads, the same amount that was spent on direct mobilization 
     and grassroots efforts.
       Seeing that the lesser part of party soft money goes toward 
     issue ads, we feel compelled to re-examine the question: how 
     is soft money harmful in elections? The obvious answer is 
     that soft money permits candidates, contributors and parties 
     to circumvent federal laws limiting campaign contributions. 
     If party soft money can help a specific candidate by using it 
     to purchase a candidate-tailored advertisement, then 
     corporations, unions or wealthy individuals can simply funnel 
     contributions to candidates through the parties. The 
     potential for the quid pro quo exchange between contributor 
     and policymaker escalates with the increasing size of 
     contributions to the party.
       But assume for a moment that party money is ``clean.'' 
     Suppose party money is generated through public subsidies, or 
     raised from contributors in increments that are small enough 
     to prevent corrupt exchanges. Are the spending patterns of 
     parties necessarily harmful in American elections? In this 
     study, we observe that parties spend a significant portion of 
     their cash to build the party as intended by the 1979 
     amendments to the FECA. It is primarily through soft money 
     that parties have had access to resources that permit them to 
     engage in activities that political scientists, for the most 
     part, view as salutary for the electoral system. If the 
     solution to the problem of corruption is to ban soft money 
     fund raising, then reformers should also consider ways to 
     ensure that parties have access to sufficient resources so 
     they might continue occupying a central role in campaigns.
       An earlier set of reforms in 1974 had the effect of 
     weakening party role in campaigns by institutionalizing PACs 
     as legitimate contributors to candidate campaigns. The number 
     of PACs proliferated in the 1970s and early 1980s, providing 
     candidates with an increasing share of their campaign funds. 
     Candidates became more reliant on PACs than on their parties, 
     which encouraged the candidate-centered nature of campaigns. 
     The ever-adaptable American parties exploited the campaign 
     finance regulations to reestablish themselves. Soft money 
     probably helped restore the party role in campaigns, making 
     the candidates less reliant on direct support from PACs. On 
     the other hand, party leaders may now feel beholden to big 
     soft money contributors, a potential problem that should not 
     be overlooked. If the soft money regime encourages interest 
     groups to contribute more frequently through the party 
     leadership, then soft money may simply centralize the corrupt 
     exchange among the most powerful political actors. If this is 
     true we should see greater party unity in congressional 
     voting than in the past, particularly for issues that are 
     important to the most generous party patrons.
       The type of party spending that concerns many campaign 
     observers is issue advertisements. In our view, party 
     spending on issue ads is not bad, per se, especially if these 
     ads link the candidate closer to party. Scholars who desire 
     responsible parties would argue that party-sponsored messages 
     create more accountability by promoting themes that unify 
     party candidates around a platform. A recent study by Krasno 
     and Seltz (2000) appears to cast doubt on this theory since 
     only 15 percent of the ads
       The fact that party money goes toward television 
     advertising reflects the reality of campaigning in a mass 
     democracy. Party leaders and their consultants believe 
     television advertising is critical to winning elections so 
     they invest in this form of campaigning. By curtailing party 
     resources, we doubt that party candidates will seek less of 
     this kind of campaign activity. In fact, reform laws that 
     cause the depletion of party resources will likely eliminate 
     ``good'' spending, such as direct voter contacts, rather than 
     ``bad'' spending, such as issue ads. Parties will employ a 
     triage strategy that emphasizes media advertising over direct 
     voter contacts and grassroots. The first activities to be 
     shorn are those that support long-term party building and 
     encourage volunteer participation, since these are not of 
     critical interest to incumbents seeking reelection.
       We also suspect that the placement of party issue ads 
     encourages electoral competition. The vast literature on 
     campaign contributions suggests that parties allocate 
     campaign resources more efficiently than interest groups, 
     preferring to give money to candidates in the closest races. 
     Interest groups tend to pursue a low risk strategy by giving 
     directly to incumbents who face little competition. Indeed, 
     parties solve a collective action problem by moving resources 
     to where they are needed most, since incumbents are often 
     unwilling to transfer money

[[Page E273]]

     from their campaign accounts to colleagues who may need it 
     more.
       Campaign resources that flow through parties, therefore, 
     will tend to promote competition more than if resources flow 
     directly into candidate committees, or when money is spent 
     independently by interest groups to promote the election of a 
     favored candidate. Using the Krasno and Seltz data for the 
     1998 elections, we observe a similar pattern of resource 
     distribution in purchasing issue ads. Table 5 (not supplied) 
     demonstrates that parties place almost 60 percent of their 
     issue ads in competitive House elections, a greater 
     percentage than either candidate committees or interest 
     groups. For Senate elections, which are much more 
     competitive, 92 percent of party issue ads appear in 
     competitive elections, whereas 74 percent all candidate-
     sponsored ads appear in competitive elections. Interest 
     groups provided less than one percent of ads in the 1998 
     Senate election, but all of these ads were placed in 
     competitive campaigns. The relatively low participation of 
     interest groups in Senate campaigns is probably because media 
     costs are prohibitively high except for the wealthiest 
     organizations.
       Candidate-controlled advertising continues to dominate the 
     airwaves, but interest groups and parties are more active 
     than ever. The only institutional counterweight to outside 
     spending by interest groups is the parties. As long as the 
     courts prevent the FEC from regulating issue ads through 
     Buckley v. Valeo, there is a danger from unilaterally 
     disarming the parties by a ban on soft money. Candidates risk 
     losing control of their campaigns in some very competitive 
     districts. Fearful of being hit by outside spending of 
     interest groups, candidates will no doubt enlist the support 
     of groups favorable to them. Indeed, there is sufficient 
     evidence in the 2000 elections that this is already 
     occurring. The groups most able to produce campaign ads for 
     candidates will likely be the wealthiest, skewing the 
     candidates' obligations toward such groups even more.
       We conclude with a policy recommendation that parties 
     retain access to sufficient campaign resources to continue 
     the activities they have pursued with soft money. Our 
     findings suggest that soft funds encourage party-building and 
     party integration, much as Congress desired when it passed 
     amendments to the campaign finance laws in 1979. To reduce 
     the potential for corruption, we recommend that Congress 
     place a cap on soft money contributions or raise the limits 
     on hard money contributions. On the other hand, we believe 
     the distinction between soft and hard money is still 
     valuable. Soft money provides an incentive for national 
     parties to transfer funds to state and local parties, where 
     campaign activities have increased substantially. We believe 
     the likelihood of grassroots work is enhanced at lower levels 
     of party, which afford more participation opportunities for 
     amateurs and volunteers. The national parties may be more 
     reluctant to transfer hard money to state parties for party 
     building when they can use this money themselves for direct 
     candidate support and issue ads.

     

                          ____________________