October 20, 2015
I applaud the commission's proposal regarding Investment Company Reporting Modernization. As a Financial Economist, I have worked extensively with the data reported by investment companies. My research using this data has been published in the Journal of Finance, Journal of Financial Economics, Review of Financial Studies, and Financial Analysts Journal. Given my extensive use of the data, I would like to suggest several changes to the current reporting requirements that would increase transparency for investors and regulators alike.
First, 40 Act funds currently report their security lending income and collateral, but there is a mismatch in the periodicity of the data. While the security lending income is measured over the whole period, the collateral is only reported as of the date of filing. Given the day-to-day structure of security loans, providing only a single day snapshot of the percentage of the fund that is on loan (or lending collateral which is a proxy for the percentage of the portfolio that is on loan, the true quantity of interest), is surprisingly uninformative. You have only to calculate the average rebate rate (security lending income divided by collateral) and compare it to industry averages to realize that a single snap-shot understates the true percentage of portfolios that are on loan. For the examples Ive looked at, the average is an order of magnitude larger than typical rebate rates for stocks on special at that time (Im happy to provide examples of this if it would be of interest).
In Question 75 of the N-SAR filings, the SEC has precedent for asking funds to report averages over time of total net assets (daily for money market funds and monthly for other funds). My guess is that TNA is a remarkably more persistent variable than lending collateral is and yet collateral is only given for a single day. The suggestion/request is that investors would have a much more complete picture of the lending activity of their funds if the collateral was averaged daily over the reporting period. Truth be told, Im surprised there isnt security-level disclosure regarding lending activity, but this suggestion has the advantage that it would require very little additional work for the investment companies relative to substantially expanded security-level disclosure.
Second, N-SARs may be filed for an individual fund or for multiple funds in a series or trust and the vast majority of funds are filed as part of a series. While many data points are reported at a fund level, even for funds in a series, there are some data points that are reported in aggregate, chiefly among these, brokerage commissions. Because brokerage commissions are paid from fund assets, the potential for agency costs regarding the payment of commissions is high. Given that these brokerage commissions ultimately have to be allocated on a fund-by-fund level in order for the proper assignment of these expenses to the differing investors across funds, requiring funds to report them at fund-level would increase transparency for investors. More generally, I would suggest that the commission revisit the rationale for any fund to file in a series. Would not investors be better served if each fund filed separately?
Third, while the current N-SAR filings require funds in a series to include their ticker, these tickers are not directly connected to the responses within the N-SAR. Each fund is identified by a different line or series number in question 7 of the filing. Unfortunately, these series numbers are not directly connected to identifying information for the funds like a ticker or a cusip. Requiring funds to report their ticker or cusip in a way that directly connects that identifying information to the fund's responses in a series filing will enable more investors and researchers to access the data and analyze its contents.
Fourth, previous research has shown that trading costs are an important determinant of fund performance. Unlike the expense ratio which is easily quantified and reported, trading costs are more difficult to standardize. I would strongly suggest that the commission think about additional trading costs disclosure to be required.
I appreciate the commission's willingness to take input on this matter. If I can be of any assistance in clarifying my comments or directing commission staff to the relevant research that supports the importance of the disclosure I have suggested above, I would be happy to do so.
Thanks,
Richard Evans
Associate Professor of Business Administration
Darden School of Business
University of Virginia