December 6, 2002

Mr. Jonathan G. Katz
Secretary, Securities and Exchange Commission
450 Fifth Street, NW
Washington, DC 20549-0609

    Re: File No. S7-3602

Dear Mr Katz

Thank you for this opportunity to comment on Proposed Rule: Disclosure of Proxy Voting Policies and Proxy Voting Records by Registered Management Investment Companies, File No. S7-36-02.

I applaud the Commission for taking this important step toward making mutual funds more accountable to investors. The project is overdue, courageous, and portentous. It holds out the hope of catalyzing further movement, voluntary and regulated, toward greater corporate governance transparency among institutional investors and other securities intermediaries. On balance, as I outlined in an article cited in the proposal release, the benefits of transparency in this area seem to outweigh the direct and indirect costs. See Palmiter, Mutual Fund Voting of Portfolio Shares: Why Not Disclose?, 23 Cardozo L. Rev. 1419 (2002).

As other comments to this proposal make clear, disclosure by mutual funds of their voting policies and records represents a significant step on a new path. As life teaches, first steps are often the most important, the most telling, the most irreversible. The Commission should strive to get this one right. The speed and direction the Commission takes on disclosure of mutual fund involvement in corporate governance will send powerful signals to other institutional investors and their regulators - at home and abroad. And where disclosure of corporate governance goes, so will go market attitudes and behavior.

Consider the Commission's 1992 initiative to require more detailed and comprehensive disclosure of executive compensation. The 1992 rules not only mandated greater disclosure of the components of executive pay, but required that the disclosure be formatted in a way that would make feasible analysis of an individual firm's pay components, as well as comparisons across firms. This greater detail has permitted the business and financial press to assemble and describe a clearer picture of executive pay. See the annual Business Week and Fortune coverage of this topic. Consistent with the rule's more detailed attention to the relationship of pay and performance, compensation committees have emphasized stock-based compensation. The more detailed disclosure has allowed academic researchers - particularly since disclosure has become available using EDGAR - to identify links (and sometimes non-links) between compensation levels/methods and firm performance.

What has been the effect of this more-detailed, uniformly-formatted, web-accessibledisclosure of executive pay? Although there was some hope that greater disclosure would shame executives and compensation committees to be less avaricious, its important effect was to provide investors a detailed (if not disturbing) picture of corporate pay practices. Important to forming this picture were information processors - particularly, the financial press and academic researchers. Detailed breakdowns of compensation levels/components and disclosure of the terms of stock-based packages were essential to identifying which compensation levels and formats seem to achieve their purposes, and which not. Significantly, most of this research focuses on periods for which EDGAR filings are available. As the dust begins to settle from the corporate scandals of the last few years, this literature provides shareholders and compensation committees a much clearer picture of what works (limited stock grants seem to) and what does not (certain stock option plans do not).

    1. Disclosure of mutual funds voting records should include data on the fund's actual votes

For mutual fund disclosure of voting policies/practices to be effective, it must be useful and accessible to informational processors - namely, the financial press and academic researchers. Just as informational processing is essential for efficient pricing by securities trading markets, informational processing is essential to the effective functioning of voting/governance markets. In fact, it is telling that numerous individual investors have submitted comments in support of the proposed rule, yet mutual funds comment that their experience has been that investors have not requested this information in the past. For individual mutual fund investors, informational processing will be critical for accomplishing the rule's avowed purpose - namely, "to enable fund shareholders to monitor their funds' involvement in the governance activities of portfolio companies, which could have a dramatic impact on shareholder value."

The proposed rule falls short in this respect. Although the proposal seems generally adequate in providing information of fund voting practices and procedures to fund investors interested in ascertaining their fund manager's voting/governance philosophy, the proposed rules mandate insufficient information to be useful to informational processors to assemble a composite of the fund's (or fund family's) actual voting record and to permit comparisons across the industry. In addition to the information proposed to be included in Item 22(b)(8), the rule should require -

  • data on how many voting shares were held by the fund as of the portfolio company's record date

  • data on how many votes were cast and how they were cast (for, against, or abstain, or withheld for particular directors)

  • uniform presentation of this data in a web-accessible, computer-downloadable format (such as a spreadsheet)

    Although this might require funds to reveal their shareholdings in portfolio companies outside the disclosure windows of the current semiannual cycle, disclosure of voting records on a semi-annual basis as part of the SAI would minimize the revelation of "fresh" competitive information. (If disclosure of shareholdings on the SAI becomes more frequent, as some haveurged, so should disclosure of voting records.) And if there exists any potential for competitive harm, the burden should be on the mutual fund industry to demonstrate that protecting this information is in the interests of the investing public.

      2. The voting record data should be presented in a uniform, web-accessible, computer-downloadable format

    To give specificity to this recommendation, voting record data should be presented as follows (15 column chart):

    1 -Issuer 2 -Ticker 3- CUSIP 4- SH meeting date 5 -Matter to be voted on 6 - SH proposal? 7- Mgmt recommend?

    XYZ Inc

    XYZI 1234568 03 Feb 03 Elect mgt slate   Yes
            Authorize new shares   Yes
            Put lawyers on board X No

    8 - Votes
    9 - Votes
    10 -Votes
    11 -Votes against 12 -Votes abstain 13 -Votes
    14 - Director withheld 15 - Consistent w/voting policies?
    456,123 456,123       456,123 Smith, Jones, Webster Yes
        456,123         No*
          456,123       Yes

    * Footnote treatment of why vote is inconsistent with fund's voting policies/practices.

    Surely, mutual funds already generate this information and likely maintain it in computer-accessible formats. If available to fund managers, the data should be available to investors (and their information intermediaries) in the same format and with the same ease.

    EDGAR has proved an excellent, accessible repository for securities information. At a minimum, voting records should be filed with the Commission and be available through EDGAR. In addition, the voting records should be available on funds' websites - if filed electronically with the Commission, the cost and difficulty of website access would seem minimal. For shareholders or other members of the public who request the voting record information, it should be made available as and when filed in printed or electronic form, as requested. Experience suggests that requests would be few.


    Some in the mutual fund industry believe that fund voting and governance activities are most effective if carried out confidentially "behind the scenes." They assert that the proposed rule would undermine these useful activities by subjecting fund managers to lobbying and pressure. (By whom is somewhat unclear.) Although I applaud the many funds (some of which I own) that are taking a proactive role in corporate governance, it is hard to fathom why the voting outcomes resulting from discussions between mutual fund managers and corporate management should be kept from mutual fund investors.

    The proposed rule would not prevent these funds from continuing practices of quiet diplomacy, or other funds from becoming more active, any more than revealing game statistics prevents a football team from calling plays in the huddle. And if, as asserted, greater transparency will harm the mutual fund game, the burden to demonstrate this should fall on those who would seek opacity. Can they sustain the burden? It seems unlikely. Consider the generally good governance results that organizations such as TIAA-CREF have achieved, even while operating with relative openness about their activities. Have these organizations been subject to untoward lobbying and pressure from their fans?

    Just as funds must disclose their investment policies and practices, disclosure of voting policies and practices assures investors that their fiduciaries are acting faithfully on their behalf. Disclosure of funds' actual voting records in a clear and accessible format will provide a metric by which to measure that faithfulness.



    Alan R. Palmiter
    Professor of Law
    Wake Forest University School of Law
    Winston-Salem, NC 27104