The SEC Brings Fund Reporting Into the 21st Century

1)    Fixing derivatives reporting
Describing a stake in IBM stock is easy. The three letters IBM, the indication that the holding is the common stock rather than one of the company’s fixed-income issues, and the number of shares (or dollar value, or percentage of fund holdings) are all that are needed. Any reasonably savvy investor will understand what to expect from that position. 

Describing a derivatives stake is another matter. For the single company IBM, there are not one but many exchange-listed futures, and not one but many exchange-listed options. Now, additional information is needed to identify the holding. The futures holding needs a date; the option needs a date and an exercise price; and both require an indication of whether the fund is long or short (short positions being rare with common stocks, but frequent with derivatives). 

Even if the fund provides all these items for its exchange-listed derivatives--which is far from universal, in fact, sometimes funds give no information whatsoever by lumping several derivatives stakes together as "Other Portfolio Holdings"--the implications are unclear. It’s one thing to see a 1.5% IBM position and evaluate what that means to the fund; it’s quite another to attempt that same exercise with a collection of IBM derivative positions. 

And those are the easy items. Things get a lot more complicated when moving off the exchanges. A customized swap might trade one flow of expected bond payments for another or exchange a fixed payment in return for a variable payment that rides with the spot price of oil. Good luck in getting the necessary information from how the swap is listed in a mutual fund report. 

Happily, the SEC is stepping into the breach. Its new proposal specifies the required data fields for each species of derivative. By themselves, those details aren’t of much help, but, when downloaded and analyzed by a computer program, they can be translated into easily understandable risk factors. Today, fund reports lack the details to make such analysis possible. Tomorrow, they will not. 

2)    Making the data accessible
Currently, funds file their portfolios electronically with the SEC. In theory, that requirement gives outside parties a single, central point for collecting information. In practice, however, those portfolios are stored in various formats, making downloading them a laborious effort. 

Recognizing the importance of data friendliness--not an issue when the ’40 Investment Act was launched--the SEC’s new proposal emphasizes "structured data formats." In other words, it specifies which data fields are to be filled, on a template that is identical across fund companies. The result should make for easy retrieval. Again, this isn’t directly of much use to investors, who are unlikely to write computer programs to scrape the data. But it greatly aids the efforts of businesses that consume the raw material of fund filings. 

Formatting will also help the SEC’s policing efforts. While the fund industry has been relatively scandal-free over the years, what problems it has suffered have tended to come from mispriced holdings. A fund company carries illiquid securities at a higher price than do its peers; the market moves against those securities; eventually, the fund company is forced to mark down its goods; and then shareholders receive a nasty surprise. The SEC would like to get in front of this pattern, rather than punish wrongdoers after the occurrence. 

3)    Publishing risk metrics
In many cases, particularly with fixed-income portfolios, professional managers base their decisions on internal statements that look very different from the officially published reports. Whereas the public reports emphasize holdings-by-holdings listing, the internal statements describe the portfolio through various "risk metrics" that summarize its overall exposure to various factors. Examples include sensitivity to changes in interest rates, credit spreads on bonds, the shape of the yield curve, and the level of stock market volatility.  

Managers are also informed about their funds' effective economic leverage, which, because of derivatives, can be significantly higher than suggested by the accounting leverage that is shown in a published fund report, and about their holdings’ liquidity. Finally, the fund might through its swaps, securities lending, and repurchase agreements have significant counterparty risk to a bank that conceivably could splat, as in Lehman in 2008. None of those items are to be found in current fund reports. 

They would, however, be found in future reports if the proposal goes through as it currently is written. Although it assumes that third-party providers will process the newly available data points to calculate their own risk metrics (as Morningstar already does with several measures--for example, average market capitalization for stock funds), the SEC also wishes that a standard, always-available set of metrics be published. 

That makes sense; while Morningstar would appreciate the business, shareholders shouldn’t need to look outside their funds’ reports to know what their money is doing. It must be granted, though, that there remain a great many devils in those details. Each risk metric may be calculated in several fashions, so if the SEC is to follow up on this particular item--which is the most ambitious of its ambitious list--then it will need to specify much investment detail.

Wrapping Up
The proposal contains other items. In penance for asking so much more from funds' operational staffs, the commission suggests that it may cut back on required mailings, permitting more filings to be done solely in electronic form. Also, the SEC asks that, in addition to maintaining the current schedule of quarterly public releases, fund companies move to a monthly frequency for a private filing that would be for the SEC's eyes only. These portfolios would provide an electronic trail for the commission's detective work.

But those, for me, are minor items. The key issues are (at long last) grappling with derivatives, making the newly required data practically useful by being easy to retrieve, and giving shareholders a portfolio manager's view with risk metrics. Any of one of those three would be good. Getting all three will be great. 

John Rekenthaler has been researching the fund industry since 1988. He is now a columnist for Morningstar.com and a member of Morningstar's investment research department. John is quick to point out that while Morningstar typically agrees with the views of the Rekenthaler Report, his views are his own.