Disclosure v. Anonymity in Campaign Finance, by Ian Ayres

[admin]

A. Details of Implementation

Mandated donor anonymity might be applied to any election. As mentioned above, some judicial election reforms have already successfully prevented candidates from learning the identity of their donors. For concreteness, this section considers how to implement a regime of mandated donor anonymity in federal elections.

[admin]

1. Private versus public administration.

One could imagine a system of literal donation booths controlled by the government: Once the curtain closed, people could drop their cash donations into a slot for the candidate of their choice, and the government would periodically pass these contributions on to the appropriate candidates. Just as there is a "ceremonial aspect[ ] of voting ... [that] is to some degree a self-conscious act of citizenship,"29 visiting a government donation booth might in time also come to be viewed as a constitutive act of citizenship.

Donation booths - whether publicly or privately administered - run greater risks of fraud than do voting booths. For either "booth" to be effective, we must trust the administrator not (1) to reveal for whom citizens vote or to whom they donate, or (2) to misapply the donation or vote to an unintended candidate. But with donations - unlike votes - there is the added risk that the administrator will convert the gift to her own private benefit.

Because of this embezzlement risk, we tentatively prefer a privatized system of blind trusts, operated by seasoned trust companies (say, those in existence for at least ten years) with substantial, preexisting assets (of more than, say, $100,000,000).30 More than 1000 financial institutions satisfy these requirements. Requiring the trust companies to be seasoned and large would make donors, candidates, and the public more likely to trust the participating institutions. The diversity of qualifying institutions would help assure that all candidates are treated fairly. But because the threat of defalcation is so high, the trusts’ records should be publicly audited ten years after each election. This ex post auditing would inform donors whether their donations had been properly routed and would allow the public to assess whether donations were - notwithstanding the trust - purchasing access or influence. Computer encryption software might make it possible for donors to verify anonymously that their contributions were credited to the appropriate campaign funds.

[admin]

2. Mechanics of blind trust operation.

Each candidate, political party, and PAC would choose a qualified institution to establish a separate blind trust account. Representatives of the blind trust could not be employed in positions influencing access or policy and, as a prophylactic, should be prohibited from privately communicating with candidates or campaign workers. The core regulation would require all donations to individual candidates, political parties, or PACs to be made to the blind trusts by mail. Campaigns would no longer be allowed to accept money in cash or by check. Campaigns would still need check books, but not deposit slips. The blind trusts would conceal the source of all contributions larger than $200. Large donors would have the option of having the trust disclose that they had given up to $200, but under no circumstance would the trust identify a donor as having contributed more than $200. Allowing donors to prove that they have contributed to a particular campaign mitigates the free speech burden of the regulation. The exact dollar amount for the anonymity threshold is unimportant, but the notion is that small donations pose a much smaller threat of corruption.

The blind trusts would then report to the candidates on a weekly or biweekly basis how much money had been donated, but would not detail the amounts given by large donors. The frequency of reporting would have to balance the candidate's need to know how much she could spend against the desire to impede candidates from decoding the identity of particular donors. Hourly disclosure of amounts available would allow a donor to say, "I bet your total went up $100,000 during the past hour." Large donations on Israel's independence day might analogously signal contributors' interest in pro-Israel policies. One way to shorten the time between disclosures would be to require that trusts intentionally obscure the presence of large donations. Trusts might even be allowed to report the daily amount available for spending, but this amount might be calculated using a randomizing procedure that breaks up unusually large contributions for future disclosure.

[admin]

3. Donor speech.

One might consider reinforcing the anonymity of the blind trust by prohibiting donors from discussing their contributions with the candidate or others. Such a prohibition could be backed up by criminal penalties, civil penalties, or both. But such a regulation is fraught with problems of enforcement and constitutionality. The law can do little to stop private, one-on-one conversations between donors and candidates. Even if such conversations could be regulated, the resulting burden on donors' free speech rights may not be compatible with the First Amendment.

A "cheap talk" regime is preferable. Just as anyone can tell Clinton they voted for him, allowing anyone to tell Clinton they gave him money - without more - would not give Clinton a very good idea of who is true contributors were. For the blind trusts to be effective, it is only necessary that donors cannot credibly communicate whether they have contributed. As long as the candidate cannot verify whether the donor's representation is true, the blind trust can impede influence peddling. Some will argue that it is simply wrong for the government to tacitly promote lying. However, it can be a civic virtue to dissemble in order to disrupt criminal activity. The possibly apocryphal World War II story of the Danish King wearing - and urging other Christians to wear - the Jewish yellow star is a prime example of the virtue of social "ambiguation."31 More prosaically, the ubiquitous (and oftentimes false) cab driver stickers - "Not more than $20 kept by driver" - shows that lying to discourage crime is an acceptable exception to truth telling.

Donors wishing to prove they donated to a particular candidate may brandish a canceled check showing the amount of their donation. To mitigate this problem, trusts should be required to providing a check cashing service for nondonors. A faux donor could mail a check to a trust with a note asking the trust to deposit the check and (once it had cleared) to mail back to the faux donor a reimbursement check. The faux donor requesting reimbursement would receive both a canceled check from her bank and a reimbursement check from the trust. A candidate seeing a canceled check made out to a blind trust couldn't be sure whether the canceled check evidences a contribution or merely a cash conversion. And since the trust's reimbursement check could be cashed or posted to a different account, showing the candidate a bank statement or audited books would not prove that a contribution had been made. As with cheap talk, appropriate regulation could undermine the credibility of canceled checks.

Donors wanting to signal their gift credibly might instead mail the check to the blind trust while in the presence of a campaign representative (or simpler yet, give the check to the campaign worker to mail to the trust on the donor's behalf). We favor prohibiting such behavior. Yet even here, a system of mandated anonymity does not need to rely solely on the deterrent effect of ex post penalties. It might be advisable to give donors a ten-day cooling-off period, during which they could cancel any donation. As long as a period exists in which donors can privately cancel their contributions, the credibility of previous public signals will be attenuated.

[admin]

4. Soliciting contributions.

The fundamental requirement would be that, in fundraising, no one from the candidate's campaign could accept contributions; only representatives of the blind trust could accept checks (via the mail). Candidates could still ask individuals for support, but they could not close the deal. Bob Dole could still have fundraisers and limit invitations to rich, registered Republicans. But under this regime of mandated anonymity, the invitations could not be conditioned on a campaign contribution, and the dinner could not be priced above cost. Instead, campaign workers could do no more than distribute postage-free envelopes addressed to the blind trust so that attendees could later mail in a contribution. Making it more difficult for candidates (and their political opponents) to solicit funds personally from wealthy contributors might alleviate the current fundraising marathon.32

This scheme of mandated anonymity would go a long way toward eliminating the longstanding practice of rewarding successful fundraisers with ambassadorships. The representatives of the trust could not take jobs or even consult with the administration. A candidate might observe a fundraiser's inputs (how many New Hampshire coffees she hosted), but not her output (how many donations she generated).

[admin]

5. Drawing the line.

In deciding what types of contributions to subject to the anonymity requirement, we will be obliged to distinguish close cases. Line drawing is a necessary feature of any reform program trying to constrain the influence of money in the political sphere. To begin, the in-kind contribution of services by political volunteers would not be made anonymously because it would be impossible for a candidate not to know their identities. Thus, people could still volunteer in order to receive undeserved access or influence. There is also no practicable way to stop candidates from knowing how much they contribute to their own campaign.33

Benefit concerts present a difficult issue. If Barbra Streisand performs a series of concerts to benefit the Clinton campaign, Clinton could easily estimate how much revenue is being generated. Allowing benefit concerts would provide an easy end run of the rule mandating that fundraising dinners must be priced at cost. Many of today's $1000-a-plate fundraising dinners could become tomorrow's $1000-a-seat benefit concerts with only nominal entertainment. Accordingly, performers should be prohibited from contractually dedicating the proceeds from an event to a political campaign. The performer or audience could independently contribute or claim that they gave or will give the proceeds; they just couldn't enter into an enforceable contract ensuring that attendance ensures contribution. We would still allow politically motivated concerts and rallies, but any profit would need to escheat to the state (or possibly to a nonpolitical charity).

[admin]

B. Can Anonymity Be Maintained?

The metaprinciple of implementation is to allow nondonors to ape easily any signal that true donors might try to send. If nondonors can mimic the signals of donors, then donors will have difficulty credibly communicating their contributions. This principle explains the specific regulations regarding donor speech, check cashing, and cooling-off periods. Instead of prohibiting donors from speaking, the regime allows nondonors to use the same words. To undermine the credibility of a donor's canceled check, the regime gives nondonors the option of acquiring an identical canceled check by merely cashing a check with the blind trust. And to undermine the credibility of mailing a check in the presence of a campaign worker, the cooling-off period allows nondonors to publicly donate and then privately cancel.

There are, however, limitations to the mimicry principle. A poor person can not credibly mimic the representations of a rich person - saying that she donated $100,000, for example. But it is unlikely that ability to pay is a close enough proxy for willingness to pay to cause politicians to kowtow to rich people generally. For example, if a law mandated that sellers of Cadillacs could not learn the identity of their customers, sellers would not respond by giving Cadillacs to the universe of rich people. Even if wealth (ability to pay) signals something about whether a donor actually gave, the important point is that the signal would be much weaker than it is now. Similarly, it would not be credible for liberals to represent that they contributed to conservatives (or vice versa). In the shadow of a donation booth, Ralph Nader could not credibly represent that he had donated to the Republican Party. At the end of the day, rich conservatives are the only people who would potentially make large soft money contributions to the Republicans. Therefore, it is reasonable to ask who among this group would be willing to go to the trouble of becoming a faux donor - to noise up the system, for example, by making the ratio of canceled checks to net donations fairly high. My answer is that the current class of Republican contributors who either feel they are being extorted or think they are paying for favors are prime candidates to fake donation. Victims of extortion are likely to have few qualms about lying to avoid the political shakedown, and even those contributors who are trying to corrupt the system by buying political favoritism may prefer to get the same favoritism for a reduced price.

Although this proposal tries to undermine a donor's ability to communicate her contribution credibly, I am under no illusion that this (or any other) system of anonymity would be completely successful in keeping candidates uninformed. Some inventive donors, with the aid of inquiring candidates, will undoubtedly devise methods to credibly signal. For example, donors or candidates may bribe a representative of the blind trust to violate her fiduciary duty and disclose donor identities.34 Undoubtedly, incumbents will have an easier time than nonincumbents discovering the identity of their contributors because a previous history of giving provides a stronger basis for belief; nonincumbents often must start with no track record of fundraising. But simply relying on reputation will not suffice. A history of giving when donations were public does not create a very strong reputation for continuing to give once contributions become anonymous. Candidates will rightfully be concerned that even faithful contributors, once behind the cloak of anonymity, will decide to chisel on their past tradition of giving.

The most predictable and serious evasions of mandated anonymity is likely to be a substitution toward "independent expenditures" or "issue advocacy." The test for what constitutes independence turns on who controls the content of the speech. Independent expenditures - in contradistinction to "coordinated expenditures" - fund political expression that is not controlled by a candidate's campaign. Independent expenditures are made without "prearrangement and coordination." The test for "issue advocacy" turns on the content of the speech itself. Issue advocacy - in contradistinction to "express advocacy" - does not expressly advocate the election of a particular candidate.

Because the Supreme Court has shown greater willingness to protect political speech that it deems either "issue advocacy" or an “independent expenditure,” mandating donor anonymity for large gifts would undoubtedly cause more extensive use of these two end runs. And it is clear that independent expenditures and issue advocacy still pose some danger of corruption. "Candidates often know who spends money on their behalf, and for this reason, an [independent] expenditure may in some contexts give rise to the same reality and appearance of corruption.”35



As shown in Figure 1, these two dichotomous categories create four permutations of control and content. Coordinated express advocacy, like candidate express advocacy, is the most regulated type of political speech. One might initially predict a hydraulic response if donor anonymity were applied to this category: Every dollar of direct contribution that the donation booth deterred might simply reemerge in one of the three other boxes - as an independent expenditure, an issue advocacy campaign, or both. Recent history has already provided ample evidence of substitution toward these three categories.36 What's more, because candidates are not accountable for "independent" ad campaigns, these campaigns are likely to be particularly negative and reckless. It is not surprising, therefore, that the infamous "Willie Horton" ads were independent expenditures.37

If mandated anonymity is likely to produce anything like a dollar-for-dollar hydraulic shift from direct contributions to independent expenditures or issue advocacy, the benefits of mandated anonymity reform would largely be lost. However, (1) mandated anonymity can be extended to reduce the possibility of an end run, and (2) where mandated anonymity is not constitutionally permissible, existing structural factors will ensure that independent or issue advocacy will not be a perfect substitute for corrupt, direct contributions. What would it mean to extend mandated anonymity? To begin, it is straightforward to cover coordinated issue advocacy. As a constitutional matter, coordinated speech can be regulated as much as direct candidate speech.38 And although there is currently a lively debate about whether current law regulates coordinated issue advocacy,39 there is little question that informational regulation (such as mandated disclosure or mandated anonymity) is constitutional.

Independent express advocacy poses a harder problem. This circumvention, however, could also be substantially reduced by requiring that such campaigns be funded solely by contributions from individuals (not corporations or unions) funneled through blind trusts.40 Under such a regime, organizations could establish committees to orchestrate independent express advocacy ad campaigns, but the funding for such campaigns would need to come from individuals' donations to blind trusts. As with the earlier anonymity proposal, individuals would be able to communicate credibly that they had contributed (up to $200) and thus, for example, have their names appear in a newspaper advertisement saying "we support candidate x." But such individuals would not be able to signal the amount of a large contribution.

Requiring that independent express advocacy be funded by individual anonymous donations would substantially reduce the viability of this circumvention. To be sure, some wealthy individuals would still be able to completely fund an independent express advocacy campaign.41 But given the costs of effective advertising, we predict that it would be difficult to raise individual contributions in the shadow of a blind trust. Those donors who are deterred by mandated anonymity from contributing directly to a candidate's campaign are unlikely to give to a blind trust that needs numerous contributions for effective independent express advocacy. And few individuals have the wherewithal to individually fund effective independent ads.

The most unyielding problem concerns substitution toward the upper right-hand box in Figure 1 - that is, substitution toward independent issue advocacy. This combination of content and control has proven constitutionally unregulable. Buckley v. Valeo suggests that mandated disclosure of speaker identity in this quadrant is unconstitutional, and mandated anonymity would fare no better. Still, some progress might be made by expanding the definition of what counts as express advocacy. The Supreme Court might accept a broader definition than the "magic words" test suggested in Buckley. The McCain-Feingold Bill attempts just this broadening by defining as express advocacy any advertisements picturing or naming a candidate within thirty days of a primary election or sixty days of a general election. But instead of capping such expenditures or requiring disclosure of the names of those people who fund such campaigns, mandating contributor anonymity would more effectively balance the government's interest in deterring corruption with the First Amendment interest in allowing unfettered discussion of political issues.

Even under the broadest imaginable constitutional definition of express advocacy, there will still be significant opportunity to use independent issue ads to affect the outcome of an election. But independent issue ads are not perfect substitutes for direct donations - especially donations made as part of quid pro quo corruption. As the Supreme Court has repeatedly emphasized:

Unlike contributions, such independent expenditures may well provide little assistance to the candidate's campaign and indeed may prove counterproductive. The absence of prearrangement and coordination of an expenditure with the candidate or his agent not only undermines the value of the expenditure to the candidate, but also alleviates the danger that expenditures will be given as a quid pro quo for improper commitments from the candidate.42

This quotation nicely underscores the procedural and substantive differences between direct contributions and independent expenditures. Procedurally, the absence of prearrangement and coordination makes it more difficult for candidates and contributors to agree on the terms of quid pro quo corruption. The inability of candidates to solicit these expenditures, in particular, is likely to reduce a candidate's ability to extort (extract rent from) potential donors. Substantively, the absence of prearrangement and coordination makes it more likely that the independent expenditure will be spent differently than the candidate would have spent a direct contribution. The Supreme Court is overly sanguine in suggesting that, "independent expenditures may well provide little assistance to the candidate's campaign and indeed may prove counterproductive." But because candidates would often use the money differently - for example, on express advocacy - candidates will tend to value $1,000,000 of independent issue ads less than $1,000,000 of direct contributions.43

Under a regime of mandated anonymity, candidates might still take positions in order to induce independent issue ads on their behalf (and vice versa), but the prohibition of both coordination and express advocacy acts as a tax on such indirect giving, tending to reduce its value to the candidate. Because mass communication exhibits dramatic economies of scale, it may be much more difficult for individuals who had been giving, say, $10,000 or $20,000 to the Democratic Party (and its candidates) to find an equally effective issue ad substitute. To be sure, independent issue ad organizations will start soliciting contributions, but these organizations are likely to find it more difficult to convince the erstwhile political donor to contribute.

While I concede mandated anonymity would lead to an increase in independent issue ads, I simultaneously predict that a regime of mandated anonymity would nevertheless reduce quid pro quo and monetary influence corruption by reducing the overall level of direct and indirect contributions - i.e., both independent expenditures and issue advocacy. A donation booth is likely to dramatically reduce the number of five and six figure “soft money” contributions. Moreover, mandated anonymity would prohibit the current practice of PAC bundling – whereby PACs gain influence with candidates by bundling together contributions from individual donors.

The predictable, hydraulic shift of contributions toward less accountable issue advocacy -- even if only partial -- is a reasonable grounds for ultimately opposing a mandated anonymity regime. But this hydraulic criticism perversely should also undermines the conviction that mandated disclosure by itself will be effective in deterring corruption. If mandated disclosure could deter corrupt direct giving, the hydraulic critics would have to fear that the same corrupt contributions would reappear as anonymous "issue advocacy" ads.44 Mandated disclosure might not deter corruption but merely shift it to less accountable independent expenditures. Proponents of mandated disclosure must admit either that finance regulation can sometimes deter unwanted direct contributions without creating an unacceptable substitution or that mandated disclosure is simply window dressing which is not really expected to deter unwanted contributions. My intuition is that the hydraulic response is not a concern when it comes to mandated disclosure because we don’t believe that disclosure deters very many direct contributions in the first place.

[admin]

C. Is the Game Worth the Candle?

This section will consider three additional drawbacks of the scheme -- beyond the shift of money to less accountable issue advocacy. While this essay has previously argued that a candidate had a legitimate interest in learning the identity of her contributors. the donation booth also denies identity information to voters and other donors. This section considers whether preventing these people from learning donor identities undermines the usefulness of mandated anonymity. But first the section considers an even more fundamental problem: whether anonymity would unduly limit a candidate's ability to speak.

[admin]

1. Less candidate speech.

The claim that mandated anonymity could cause a campaign-financing crisis must be taken seriously. Anonymity exacerbates the donor's paradox for large donors and might lead to a dramatic drop-off in giving. As a general matter, donors like to be recognized for their charity. The donation booth may have an overbreadth problem in that contributors who currently give, in part, to acquire status among their peers may be deterred from giving through blind trusts. Even donors who are not motivated by a desire to corruptly influence policy may thus be chilled by mandated anonymity.

Access to the media requires funding. A reduction in donations could mean a reduction in media access. In this regard, mandated anonymity could limit a candidate's ability to speak - and the public's right to listen. Indeed, the very uncertainty of the effect of mandated anonymity on contributions could give policymakers pause.

A related concern is that, by reducing the ability of candidates to speak, mandated anonymity will unduly increase the influence of other speakers, such as the media, unions, and rich, self-funded candidates. Media speech, the quintessential independent expenditure, will go unregulated under any reform proposal. We might worry about who will be next in line to influence the candidate corruptly if anonymity undermines the influence of large donors. Candidates unable to sell influence in exchange for contributions might begin to kowtow to the imagemakers of the mass media. It might be better to countenance the undue influence of large donors under the current system than to transfer this influence to an even smaller media oligarchy. Under this theory, the contributions of James Riady and the millions of other millionaires among us may provide a Jeffersonian counterweight against the potentially disproportionate influence of Citizens Hearst or Murdoch - or the even less accountable corporations and unions that bankroll issue ads.

Nevertheless, facilitating quid pro quo and monetary influence corruption is too high a price to pay for political speech. The Constitution doesn't require Congress to facilitate corruption in order to subsidize political speech. Prohibiting quid pro quo deals might also substantially reduce the ability of candidates to speak, but the First Amendment doesn't mandate generating money to produce a meaningless debate in which donors have already purchased candidates' positions outside the realm of open deliberation. If noncorrupt private donations do not sufficiently fund campaigns or offset the undue influence of media moguls, we should supplement private contributions with public money. A belief that mandated anonymity would produce far fewer net political expenditures than a mandated disclosure should be a signal that disclosure by itself would not be effective in deterring corruption.

[admin]

2. Less donor information for voters.

Mandated anonymity keeps voters - as well as candidates - in the dark about donors' identities. Denying voters this information could be problematic. The Supreme Court in Buckley identified two adverse effects:

[Disclosing the identity of a candidate's donors] allows voters to place each candidate in the political spectrum more precisely than is often possible solely on the basis of party labels and campaign speeches. The sources of a candidate's financial support also alert the voter to the interests to which a candidate is most likely to be responsive and thus facilitate predictions of future performance in office.45

The second advantage of donor identity is absent under a system of mandated anonymity: Candidates are not "more likely to be responsive" to donors if they don't know who their donors are. Moreover, it is unclear whether the first effect of donor identification - more precisely placing the candidate in the political spectrum - should be classified as an advantage. It might be more conducive to democratic deliberation for voters to learn about a candidate's positions on policy matters rather than to learn whether Jane Fonda or the NRA contributed to the candidate's campaign. Individual donors at times may have better information - possibly based on private conversations with the candidates - about a candidate's true intentions than could be gleaned from the public record. But because there are other avenues of gaining this information, the government's interest in contributor identity as a proxy for candidate beliefs is less compelling than its interest in deterring corruption.

The proposed regime of mandated anonymity also partially accommodates the voters' interest in donor identity. Under the proposal, a contributor would have the option of having the blind trust disclose the amount of her contribution up to $200. Given the pervasive interest of donors in identifying themselves, it is likely that the vast majority of donors would opt to be identified as having given something. Given that voter knowledge of donor identity is less important in a regime in which candidates as well as voters are kept in the dark, and given that some voter information about donor identity would be generated under the proposed system of optional partial disclosure, the public's interest in donation information does not ultimately militate against the proposed anonymity regime.