Answering Ayres: Requiring campaign donors to remain anonymous would not resolve corruption concerns. (Campaign Finance).

Mayer, Kenneth R.

IN A RECENT REGULATION ARTICLE, YALE LAW School professor Ian Ayres offered a radical solution to the perpetual controversy over campaign finance reform ("Should Campaign Donors Be Identified?" Summer 2000). Recapitulating a proposal that he (along with economist Jeremy Bulow) first set out in a 1998 stanford Law Review article, Ayres suggests that all campaign contributions be anonymous, so that candidates would not know the identity of their contributors or, at least, not know the amount that each contributor gave. Such a restriction of information would cripple efforts to trade access and influence for campaign contributions, he believes.

The "donation booth" model, as Ayres calls his proposal, has not gained much attention in mainstream society, but it has received considerable support from some legal scholars and pundits. Even Federal Elections Commission (FEC) member Bradley Smith, who is no great fan of campaign finance regulation, has praised the concept as "a type of creative thinking that may deserve more attention."

Ayres' blind trust strategy is worth consideration because it offers an innovative solution to what appears to be an intractable problem. However, after unpacking the donation booth's assumptions and analyzing its implications, I believe observers will see that it is a remarkably bad idea that reflects some fundamental misunderstandings about the political process.

AYRES' PROPOSAL

Instead of allowing candidates to raise money directly, Ayres proposes that they be required to establish a blind trust for campaign fundraising, through which all campaign contributions would be filtered. Candidates would not be informed of who gave what amount, but instead would receive only a general contribution total and a simple list of campaign donors. The only exception that Ayres would allow is for the trust manager to denote whether contributors gave $200 or more, but the candidate would not be informed of how much more than $200 the contributors gave. Only after a lengthy period of time (Ayres suggests 10 years) would the records showing contributor names and donation amounts be made public, at which point they would be audited and available for public inspection.

Cheap talk Under the donation booth scheme, contributors could still claim to candidates that they had donated a certain amount of money, but they would not be able to prove it. Ayres' concept includes a few wrinkles to further reduce the credibility of such claims: Donors would be able to ask for a refund of their contributions within a specified cooling-off period (so a canceled check would not be sufficient proof), and large donations would be broken up into random-size chunks and distributed to candidates over time (so that an especially large donation would not create a noticeable spike in what a candidate receives).

Those provisions would create what Ayres calls a "cheap talk" regime in which non-donors could claim that they gave money (and expect to receive favorable treatment). But candidates would not be able to lavish official favors or sell access to wealthy donors, because they would never be sure who had contributed what. And donors would no longer be assured of favorable treatment because they could not prove that they had really given to office holders' campaigns.

Anonymity and corruption The donation booth, Ayres claims, is analogous to another type of enforced anonymity created to interfere with political corruption: the voting booth. The secret ballot eliminated widespread vote buying and intimidation after its introduction in the late nineteenth century -- once voters cast their ballot privately, party leaders could no longer be sure who really voted the ticket, and voters could no longer prove that they had voted as instructed. Absent proof, vote-buying agreements were, by definition, unenforceable. So too, the argument goes, would the donation booth interfere with the market for political influence, because any influence-peddling agreement would be similarly unenforceable.

CHALLENGING AYRES

Objections to the donation booth concept are both philosophical and practical, and fall into three general categories:

* The comparison between the donation booth and the voting booth is flawed because the secret ballot is an individual right that a person can waive, not an affirmative obligation that requires an individual to keep his vote secret. Ayres fails to note that ballot secrecy, far from being inviolate, is compromised routinely in ways that allow people to prove how they voted.

* The donation booth would not interfere with credible donation claims because donors would have little incentive to lie and every reason to tell the truth. Even if Ayres is correct that a cheap talk regime would work, surely one need hardly point out the irony of a political reform designed to raise the level of political debate by providing an institutionalized incentive to lie.

* A blind trust would require the government to prohibit the release of donation records, which would create an astonishing and unprecedented category of state secret, with the government forced to apply sanctions to anyone who chooses to reveal it.

Let us look more carefully at those three objections.

ANONYMOUS CONTRIBUTIONS AND SECRET BALLOTS

The primary philosophical justification advanced for the donation booth concept is that we cast our ballots under a similar veil of anonymity. Since the introduction of the Australian ballot in the 1880s, virtually all voting is conducted secretly. We have no verifiable record of how we vote, so we cannot prove to anyone that we actually cast our ballots one way or another.

A key -- though certainly not the only -- cause of the ballot reform movement was the desire to stamp out the corruption and intimidation that occurred when voters cast ballots under the watchful eyes of party regulars. Political parties produced separate ballots that they distributed to voters outside polling locations; the ballots were printed to make it easy to see how people voted. The abuses of the time have been well documented: Votes were often bought outright, and voters faced intimidation by employers and party officials or threats from violent mobs. The secret ballot was an ideal solution to those problems. Once political parties could not observe how individuals voted, the incentive for corrupt bargains disappeared.

No anonymity requirement Why not, the thinking goes, extend that solution to cover campaign finance corruption? Both the secret ballot and Ayres' donation booth target "quid pro quo" corruption; the only difference is what is for sale. The secret ballot prevents parties and candidates from purchasing the official act of an individual voter, while the donation booth prevents contributors from purchasing the official act of an elected official. If the secret ballot diminished vote buying and intimidation directed at voters, why would secret contributions not have the same effect on candidates? "Anyone opposing mandated anonymity," wrote Ayres in his Regulation article, "needs to explain why we should not also jettison mandated voting anonymity."

Fair enough. Here goes: There is no mandated voting anonymity.

In repeatedly defending the donation booth as a close cousin to the voting booth, Ayres misses the crucial point that the secret ballot is a personal right, not a state-imposed obligation. That is, the state may not compel me to reveal my vote, but neither may it prevent me from doing so. Decades of state case law have established that principle. One of the more frequently cited cases, a 1920 decision by the North Carolina Supreme Court (Jenkins v. State Board of Elections of North Carolina), was emphatic on that point, concluding,

[The] privilege of voting a secret ballot has been held to be entirely a personal one. The provision has been generally adopted in this country for the protection of the voter, and for the preservation of his independence, in the exercise of this most important franchise. But he has the right to waive his privilege and testify to the contents of his ballot The voter has the right at the time of voting voluntarily to make public his ballot, and its contents in such case may be proven by the testimony of those who are present.

It is true that many state election codes include language that, on its face, prohibits voters from revealing their ballots after casting them. But there are many exceptions, and the fact that a voter has either voluntarily or mistakenly exposed her ballot, by itself, is almost never sufficient reason to prosecute or invalidate her vote. Oregon, which now conducts its elections entirely by mail, requires voters to place their ballots in secrecy envelopes before sending them to election officials. Nevertheless, votes are still counted even if voters fail to abide by the rule.

No secrecy In some circumstances, the ballot is not secret at all. Under federal law, voters with disabilities, or those who are illiterate, are permitted to have assistance from a person of their own choosing, so long as that person is not the voter's employer or union representative. The Arkansas state constitution requires paper ballots to be numbered, in order to permit election officials to connect specific ballots to individual voters if an election is contested. In closed primaries where only registered members of a political party are eligible to vote, voters must reveal their partisan preferences before receiving a ballot.

Courts have long noted the lack of secrecy inherent in absentee balloting, but none has ever held that the lack, in itself, violates constitutional or statutory provisions that mandate secret ballots. The Florida Supreme Court, in Hutchins v. Tucker, held that an absentee voter has waived his right to a secret ballot altogether.

What is more, as a rule, voters have every right to show their completed absentee ballots to just about anyone. Again, clear evidence of that proposition comes from state case law (in this instance, a 1982 California Appeals Court decision, Beatie v. Davila):

[If] a voter wishes to disclose his marked ballot to someone else, be it a family member, friend or a candidate's representative, he should be permitted to do so.... We suspect that many absentee voters disclose their marked ballots to other persons before placing them in the identification envelope for return to the elections official or the polling place. Such a voluntary disclosure cannot be deemed to violate the constitutional mandate [of secret ballots].

The transparency of the voting booth under such circumstances is hardly trivial, given the increasing popularity of absentee balloting and mail-in votes. In the 2000 election, over 2.7 million Californians voted absentee, and 1.56 million Oregon voters cast mail-in ballots. In the state of Washington, nearly half of all ballots were cast absentee.

Ayres' secret-ballot comparison is thus more rhetorical than analytical, relying as it does on a highly stylized and inaccurate picture of the voting booth. Voters, it turns out, cast their ballots in a variety of circumstances, with the degree of ballot secrecy dependent almost entirely on what the individual voter chooses to keep secret. Ayres thus misidentifies the nature of the right to a secret ballot: It vests with the individual voter, not the state. In his Regulation article, Ayres conflates mandated anonymity with "the cherished freedom not to speak," although the two are not the same. The voting booth/donation booth analogy obliterates the distinction between "I have no obligation" and "I am prohibited."

Ballots vs. campaigns Even if Ayres is correct in asserting that the secret ballot is a form of mandated anonymity, his comparison to the donation booth is still flawed because the specific act of casting a vote is very different from what goes on in the far more amorphous and broader campaign environment. Hence, courts have permitted states to regulate the voting process in ways that would be patently unconstitutional if applied to the campaign process.

In campaigns, voters assimilate the information they want in order to make the decision about which candidate to support. Candidates attempt to mobilize their supporters, persuade the undecided, attack each other, and make promises about what they intend to do once elected. Campaigns are often about artifice, rhetoric, spin, and even outright falsehoods. Rather than police those activities, lawmakers and the courts have decided to leave it to the voters to sort through everything. Voting, in contrast, involves the discrete mechanism of choice, the specific administrative processes, and the counting rules that determine the winner of a particular contest.

We thus draw distinctions between political/campaign rhetoric, which generally is unregulated as part of the "marketplace of ideas," and the physical process of marking and submitting a ballot, which is controlled far more carefully. A congressional candidate who offers to send a $20 bill to every voter in the district if he is elected likely would be charged with vote fraud. A candidate in the same race who proposes a tax cut of $1,000 per household would be guilty of nothing more than making a campaign promise.

Several U.S. Supreme Court decisions have recognized that the polls are distinct, conceptually and practically, from what goes on outside. In Mills v. Alabama (1966), the Court invalidated an Alabama law making it a crime to solicit votes on Election Day; a newspaper had been prosecuted for publishing an election-day editorial that made election endorsements. "It is difficult to conceive of a more obvious and flagrant abridgment of the constitutionally guaranteed freedom of the press," wrote Justice Hugo Black. Yet in Burson v. Freeman (1992) the Court upheld a Tennessee law barring campaign activity within 100 feet of a polling place. Protection of the integrity of elections and prevention of voter intimidation were seen as compelling state interests that justified speech limits within a narrow physical zone in and around the polls. The two rulings indicate the Court's willingness to place restrictions on activities where voting is actually occurring, but its refusal to impose similar restrictions on campaig n-related activity in general society.

The state can thus exercise much more control over the voting process than it can over the campaign process. Restrictions on political speech that are permissible inside and around the polling place are not permissible beyond that. Consequently, even if it were established beyond a doubt that the secret ballot was mandatory, it does not follow that similar restrictions would be tolerable in other contexts.

IS TALK REALLY CHEAP?

A -- perhaps the -- foundation of the blind trust concept is that non-donors must be able to behave exactly like donors, so that candidates can never be certain who really gave them money and who -- to put it bluntly -- lied. If potential donors want to receive the benefits of contributing -- access, favorable legislation, etc.--without actually bearing the cost of donating, they simply could claim that they have given. In the presence of such free-riding, and without proof one way or the other, candidates would have no way to be certain which claims were true and which were not.

The blind trust can only work if candidates are sufficiently confused by Ayres' "cheap talk" regime. But both theory and practice suggest that candidates will not face such confusion. Even under an anonymous system, committed donors would still have incentives to be honest with candidates, and non-donors would face compelling disincentives against misrepresentation. In a political environment where ongoing relationships are vital to both candidates and interest groups, the desire to be seen as credible would, over the long term, provide a powerful reason for interest groups to tell the truth.

Honesty and politics Cheap talk falls into two categories: claims made by people who purposely are trying to inject confusing information into the election, and those made by otherwise credible contributors who want to obtain influence without actually spending any money. Legislators and candidates are unlikely to be confused by the first type of false claim, although voters might be. It is the second type of false claim that is more interesting, and more likely. How plausible is that expectation? How would a candidate assess the credibility of a claim, either public or private, that a donor gave $10,000?

Consider that everyone would know that, after 10 years (or however long a period must elapse before contribution records are released), a false statement is certain to be exposed. Some false contributors would be discovered even earlier: A person who claimed to have given $100,000 to a candidate who only raised $50,000 over an entire election cycle would be exposed quickly as a liar. When a false claim becomes known, the people who made it would see their credibility drop to zero in all future interactions. Nobody with long-term stakes in the political process would be willing to risk that. Even donors who want to send confusing signals would benefit from a longstanding pattern of honesty, because that reputation would enhance the credibility of a false claim if it were ever made, and even a single false claim would raise doubts about that donor's true activities.

The cheap talk regime, then, assumes that there are no reputational consequences to lying. That notion is contradicted by overwhelming evidence that virtually everyone involved in striking political bargains is acutely aware of the importance of reputation. Indeed, reputation is central to political deals precisely because most are unenforceable if one party fails to uphold its end of the bargain.

Weak interest groups? Conceivably, the cheap talk regime might protect wealthy contributors from politicians who try to squeeze them into making large donations. Economists refer to that as "rent extraction" behavior, although most people would call it "extortion." Corporations or other large contributors may feel strong pressure to ante up to political parties or influential legislators out of fear of what might happen if they do not. Victims of such demands, Ayres has argued, are especially likely to claim falsely that they have donated.

In claiming that organized interests need protection against legislative arm-twisting, Ayres has flipped the legal rationale for campaign finance reform on its head. No longer is regulation justified in order to prevent wealthy donors from purchasing access or favors (the "corruption or appearance of corruption" rationale accepted by Buckley as the only compelling interest that justifies regulation), but reform also is necessary to protect the wealthy and highly organized.

That is a very strange argument for Ayres to place alongside the broader claim that the existing system grants the wealthy disproportionate political influence. In doing so, he has constructed a model of politics in which wealthy interest groups are both so powerful that they can get whatever they want out of Congress, and so weak that they are at the mercy of Congress if they do not do what vengeful legislators ask. Although either condition could easily be considered corrupt, both cannot be true simultaneously.

Information cues In any event, the most likely consequence of the blind trust regime is that it would deprive the public, but not the candidates, of information regarding donations. Large donors would simply make their claims privately to candidates. The private signals would still be credible; candidates would still know with reasonable certainty who is giving to them, but the public would be in the dark. Currently, even though the public might not know what interest groups say to candidates, we do know what they give. Under the blind trust regime, neither communications nor contributions would be publicly visible. That hardly seems an improvement over the existing regime.

Moreover, to the extent that the cheap talk regime works, it would have the effect of driving valuable information out of the electoral arena. If candidates would be confused about the identity of their real contributors, then it is axiomatic that voters would be even more so. The potential advantages in disrupting potential corruption must therefore be balanced against the disadvantages in forcing voters to make choices without important cues.

A significant literature attests to the importance of information shortcuts, or cues, that voters use to evaluate candidates when detailed information is hard to come by or absorb. One of the most valued cues is the identity of a candidate's supporters. Such cues allow otherwise unsophisticated and uninformed voters to act rationally, in the sense that they can make the same decisions that they would have made if they had invested considerable time in investigating the issues. Interest groups have the incentive and ability to evaluate political information about a candidate's preferences, behavior, history, and policy positions. A group's willingness to commit its resources to a candidate is an unambiguous signal of where it stands, and knowledge about who has given money can be vital to a voter's evaluation of a candidate: Did she raise her funds from Emily's List, or the Conservative Action Fund? Handgun Control or the NRA Political Victory Fund? Driving that information out of the campaign, or intentionall y introducing confusion, seems unlikely to improve the election process.

A NEW STATE SECRET?

No matter how much attention is devoted to the process of maintaining the confidentiality of contribution data, some unambiguous, traceable, and authoritative record of the origin, amount, and destination of each contribution would have to exist, even under Ayres' scheme. The need for such definitive records is obvious, as without them it would be impossible to carry out the audits necessary to insure that contributions go where the donors intended, that contribution limits are followed, and that other important restrictions (such as bans on foreign and corporate contributions) are enforced. Thousands of people would need routine access to that information, and the blind trust system would have to mesh closely with existing financial institutions to insure that contributions are debited properly from donor accounts.

It is thus a given that, even under Ayres' system, the information would exist, and no encryption code or organizational wall could keep names separate from dollar amounts forever. How, then, could regulators keep it from being revealed? Ayres' solution is to require that trust employees not be employed as lobbyists, not have private contact with candidates or campaign staff, and that "under no circumstances" would the amount given by a donor be revealed (other than that it was either above or below $200).

Enforcement But it is not at all clear how such rules could be enforced, and even less clear that they would work. Ultimately, any blind trust regime would have to impose some sort of sanction against trust individuals who disclosed official documents concerning contributions. Otherwise, it is inevitable that those with access to the information would find a way to transmit it to candidates or the public; at the extreme, trust employees could simply provide documents openly to reporters or candidates, or selectively disclose particular contributions through leaks. The dismal track record of restrictions on the dissemination of political information -- e.g., leaks of Voter News Service exit poll data, total flouting of the French ban on the broadcast of public opinion polls within two weeks of elections -- should engender considerable skepticism that contribution records would be kept under wraps.

The possibility of private traffic in contribution data leads to all manner of pathologies. Candidates could seek damaging information about their opponents in the hope of exposing contributions from out-of-favor interests or individuals. Alternatively, they simply could assert that their opponents have taken tainted money, and the same restrictions that prevent a candidate from knowing who has given her money would prevent her from proving that she had not received suspect funds. Absent full disclosure, the public would be denied the information necessary to evaluate the competing claims.

Punishment What sanctions would be permissible to prevent leaks and unauthorized releases? Ayres is silent on that question in his Regulation article, although elsewhere he has suggested that trustees would have a "fiduciary duty" to maintain confidentiality, and proposed that the FEC be authorized to conduct field tests to see if campaigns and trusts are willing to compromise donor anonymity.

But Ayres' investigative mechanisms and assertions of fiduciary responsibility finesse the issue of enforcement. One reason, perhaps, for not confronting it head-on is that there is neither a coherent legal foundation nor any precedent for punishing people who would disclose campaign contribution data. In order to implement the donation booth, the federal government would have to create a completely new category of government secret: data on how much individuals contributed to federal candidates. Restrictions on disclosure of that top-secret information would have to be backed up with some form of penalty, whether it involved dismissing a loose-lipped trust employee or launching a criminal prosecution.

It is difficult to imagine how such a restriction would pass First Amendment muster. If the United States government could not successfully prosecute Daniel Ellsberg for leaking the Pentagon Papers to the New York Times, it is hard to see how it could punish a trust employee for telling a reporter how much money Emily's List passed along to Barbara Boxer. How it could prevent a former employee (who could no longer be fired from a position in the trust) from revealing such information is even harder to fathom.

No category of information currently protected from unauthorized disclosure -- whether we are speaking of nuclear weapons data or autopsy photographs of NASCAR driver Dale Earnhardt -- provides any support to the donation booth concept of coerced secrecy. With every other type of privileged information, disclosure would violate some individual privacy right, compromise a property or proprietary interest, or interfere with a vital government function such as national security or law enforcement. Nothing of the sort applies to the donation booth, in which the parties to a particular contribution would, presumably, want the information released. In no way can the consequences of releasing contribution data be compared to the harm in releasing information on a law enforcement investigation or a diplomatic initiative, nor can it be claimed that release would interfere with efficient governmental administration. The government cannot assert a legitimate (let alone compelling) interest in keeping such information co nfidential.

Ayres' insistence that a donor's ability to say whatever she wants mitigates the secrecy problem misidentifies the location of the constitutional offense. It is not a question of whether or not a donor has the right to "prove" that she did make a contribution. It is, rather, two separate questions: By what authority can the government prohibit the ability to offer that proof? And, by what authority can the government prohibit a third party from releasing information on candidate or donor activity if the third party has access to that information and is not restricted by contract from doing so?

CONCLUSION

Ayres and others have advanced the donation booth concept as a novel way of disrupting quid pro quo corruption while safeguarding individual rights, just as the secret ballot cut down on election fraud and protected voters from intimidation. But, as my analysis has shown, the blind trust is based on a flawed comparison with the secret ballot. It would not interfere with credible signals about donation activity, though it would allow donors to send those signals privately. And, it would require the government to treat contribution records as state secrets.

One surprising aspect of Ayres' proposal is how little criticism it has generated. How can such an idea be taken seriously? One possibility is that the donation booth reflects a common characteristic of much of the contemporary legal literature on campaign finance, which is deeply suspicious of "politics" as a legitimate system for making collective decisions. Politics, when viewed through that lens, inevitably corrupts what would otherwise be a rational and consensual process, diverting official judgments away from what everyone should recognize as the common good. That conceptual framework justifies attempts to impose a particular kind of normative rationality on politics and campaigns, in which voters and candidates set aside all bias and deliberate solely on the basis of policy rather than self-interest.

But the framework is built on a flawed foundation that very nearly denies the possibility of sincere policy preferences or the validity of expressing philosophical support or opposition through a campaign donation. Similarly, the donation booth is based on the unjustifiable notion that state-imposed secrecy would improve the political process.

READINGS

* "Commital Deniable Proofs and Electronic Campaign Finance," by Matt Franklin and Tomas Sander. ASIACRYPT 2000, available online at citeseer.nj.nec.com/439710.html.

* "The Donation Booth: Mandating Donor Anonymity to Disrupt the Market for Political Influence," by Ian Ayres and Jeremy Bulow. Stanford Law Review, Vol. 50 (1998).

* "Moralism and Realism in Campaign Finance Reform," by Bruce E. Cain. University of Chicago Legal Forum, Vol. 111 (1995).

* "Paying for Politics," by Geoffrey Brennan and Alan Hamlin. Designing Democratic Institutions, edited by Ian Shapiro and Stephen Macedo. New York, N.Y.: NYU Press, 2000.

* "Shortcuts Versus Encyclopedias: Information and Voting Behavior in California Insurance Reform Elections," by Arthur Lupia. American Political Science Review, Vol. 88 (1994).

* "Should Campaign Donors Be Identified?" by Ian Ayres. Regulation, Vol. 24, No. 2 (Summer 2001).

* "A Theory of Credibility," by Joel Sobel. The Review of Economic Studies, Vol. 52 (1985).

* Unfree Speech: The Folly of Campaign Finance Reform, by Bradley A. Smith. Princeton, N.J.: Princeton University Press, 2001.

Kenneth R. Mayer is a professor of political science at the University of Wisconsin, Madison, where he specializes in such areas as defense policy, campaign finance, and the presidency. Mayer is the author of several books, including With the Stroke of a Pen: Executive Orders and Presidential Power and The Political Economy of Defense Contracting. He can be contacted by e-mail at kmayer@polisci.wisc.edu.