Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

In this Discussion

Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.

    Support MFO

  • Donate through PayPal

Calling EDGAR experts at MFO

Suppose I want to find the year by year expense ratio for some mutual fund of interest--in this example, the Vanguard 500 Index fund, originally VFINX. Easy enough to get its performance from M* etc. back to its beginning in 1976, and easy enough to get the S&P index performance over that period from the SBBI. Subtracting the two gives an estimate of expenses, but only an estimate; strictly speaking, the subtraction gives tracking error, not expenses incurred.

Like all funds, it files at the SEC and these filings go into the EDGAR database. Naively, since EDGAR goes back to 1994 (I am told), I hoped to get those expense ratios from EDGAR, and maybe, in my dreams, find a 10-year trailing table back to 1984 from the 1994 filing. Add seven years of Wiesenberger yearbooks and I'd be done.

But when I enter the ticker at the EDGAR search engine, 2013 is the oldest report listed. However, that search was entered at the "front page" and I am a naïve EDGAR user.

Some members here will not be naïve EDGAR users. Do you know how to get a pre-2013 filing on VFINX online? Or where else might you send me for that year by year data on expense ratios? (I have John Bogle's data on expense ratios for Vanguard as a whole, but that doesn't answer my question)

Why do I care? Tracking error can be positive, due to security lending etc. Only with a separate calculation of expense can one vet how well fund management did given the expenses they were dealt. Not so important with an index fund, maybe; more so with the ordinary sort of fund.

PS: were this data to be available in MFO Premium, please tell me!

Comments

  • msf
    edited July 2022
    I think this is the result you got.
    https://www.sec.gov/edgar/browse/?CIK=0000036405

    Try switching to the "classic" version. As M* and Coke have demonstrated, "new" is not necessarily improved.


    This "SEC classic" page links to filings all the way back to 1994.
    https://www.sec.gov/cgi-bin/browse-edgar?CIK=0000036405&owner=exclude

    Until 1998 the fund was known as Vanguard Index Trust 500 Portfolio.
    https://www.wsj.com/articles/SB852585751767090500
  • There is a "classic" version of EDGAR? Who knew?

    Love the dig at New Coke. A staple of branding lectures for years when I started teaching in the 1980s.

    Thanks again, msf! And thanks to everyone who keeps this board going, it has already proved quite the resource.
    msf said:

    I think this is the result you got.
    https://www.sec.gov/edgar/browse/?CIK=0000036405

    Try switching to the "classic" version. As M* and Coke have demonstrated, "new" is not necessarily improved.


    This "SEC classic" page links to filings all the way back to 1994.
    https://www.sec.gov/cgi-bin/browse-edgar?CIK=0000036405&owner=exclude

    Until 1998 the fund was known as Vanguard Index Trust 500 Portfolio.
    https://www.wsj.com/articles/SB852585751767090500

  • As a further update, and in the spirit of leaving bread crumbs behind for anyone else who wants to attempt such a search, on some other mutual fund:
    -Even classic EDGAR stops about 2001, limited to trailing 20 years at a guess. Even entering the old name ("Vanguard Index Trust") doesn't pull up anything earlier.
    -Casting around, I found this page: https://www.sec.gov/edgar/search-and-access. At the very bottom of the page is a link to EDGAR archives. That sounded promising. It took me to this page: https://www.sec.gov/cgi-bin/srch-edgar. Bingo: searching here for Vanguard 1994 to 2000 brought up the annual reports from the 1990s. (N-30D is the code for AR)
    -"Financial highlights" has the prior five years of expense ratios. Another tidbit: in the 1990s the norm, at least for Vanguard, was to present twenty years of trailing returns; so in the 1996 report one can see the 500 Index fund annual returns back to its inception in 1976. Will be interesting to see if other fund firms did the same. (The 500 returns are on the Simba backtesting spreadsheet at Bogleheads.org, so no treasure hunt in its case).
  • edited July 2022
    Interesting that if one backtracks to general search page from either the new/current or Classic Edgar (and even the link provided by @msf), one gets to the same search page/URL:

    https://www.sec.gov/edgar/searchedgar/companysearch.html

    https://www.sec.gov/edgar/searchedgar/companysearch.html


    The historical/archival search link noted by @mcq seems to be available only from Edgar – Search & Access click (on the left menu)

    https://www.sec.gov/cgi-bin/srch-edgar

    “Welcome to the archive of historical EDGAR documents. This search allows you to enter complex queries to retrieve all but the most recent day's EDGAR filings (from 1994 through 2022). If a simple search will suffice or if you need real-time, up-to-the-minute filings, please visit the main EDGAR Search page for other choices.”


    It seems that Edgar has evolved over the years and some features are buried deeply within other clicks/tabs.
  • msf
    edited July 2022
    Reiterating what I did:

    - Starting with mcq's search result, obtained by searching for VFINX on the mutual fund search page and following the CIK 0000036405 link on the results page for the Vanguard Index Funds trust ...

    - Follow the "Classic version" link in the upper right corner to the classic result page.

    Another search for the trust or fund is not needed. One has already navigated to the classic version of the results page. The only search remaining to do (for convenience) is to limit the results to the particular forms desired. One could instead scroll through all the results, all the way back to 1994.

    For annual (N-CSR) and semi-annual (N-CSRS) reports from 2003 and later, one can click the "Shareholder reports" button. This autofills the "Filing type" to be N-CSR (which matches both forms), and submits the query.

    For annual and semi-annual reports reports prior to 2004, the SEC required investment companies to use form N-30D. Manually fill the form box with N-30D. A search returns the reports between 1994 and 2003.
    Once you [RIC] begin filing Form N-CSR, you will no longer submit shareholder reports on EDGAR as N-30D submissions. Instead, you will include your shareholder report as part of your N-CSR submission.
    Frequently Asked Questions: EDGAR Filing of Certified Shareholder Reports by Registered Management Investment Companies

    the annual reports from the 1990s. (N-30D is the code for AR)
    Exactly.
  • Arbitrating between yogibearbull and mfs, there may be a browser interaction issue.

    When I clicked on the middle link in mfs' initial reply:

    This "SEC classic" page links to filings all the way back to 1994.
    https://www.sec.gov/cgi-bin/browse-edgar?CIK=0000036405&owner=exclude


    I landed here:
    image

    Nothing about classic in the upper right corner. Maybe because I was using Chrome on a Windows 11 machine?? Who knows.

    Was my "archives" link found also by yogibearbull actually a link to EDGAR classic? I don't know.

    Either way, I'm grateful for the advice. It was rather entertaining to read John Bogle's shareholder letters back in the 1990s.

    Another tidbit: the V long investment grade bond fund was the "long corporate bond" fund in reports from the 1990s; and it got its start in 1973 as the Westminster Bond fund.

    Wellington. Wellesley. Westminster. Anybody else see a pattern here:-)
  • Going into even more detail ...

    Step 1:
    Orig post: I think this is the result you got.
    https://www.sec.gov/edgar/browse/?CIK=0000036405

    New post: Starting with mcq's search result

    Embedded link in new post is identical to URL given in orig post. It brings you to the page displayed below. Note the pre-filled starting date of 2013-08-12 (as far back as the new Edgar goes). Also note the "Classic version" link in the upper righthand corner.

    image

    Step 2:
    New post: Follow the "Classic version" link in the upper right corner to the classic result page

    Orig post: This "SEC classic" page ...
    https://www.sec.gov/cgi-bin/browse-edgar?CIK=0000036405&owner=exclude

    Embedded link in new post is identical to URL given in orig post. It brings you to the classic version of the results page.

    image

    Step 3:
    Orig post: ... links to filings all the way back to 1994

    New post: One has already navigated to the classic version of the results page. The only search remaining to do (for convenience) is to limit the results to the particular forms desired. One could instead scroll through all the results, all the way back to 1994.

    -----

    The links to the filings on this "classic version" results page are the "Documents" buttons. To "scroll" through all the results, one scrolls through the first 40 results and at the bottom of the page, clicks on the "Next 40" button to get the next 40 results, and so on. This takes you back to increasingly older filings until one reaches the 1994 filings.

    Instead of scrolling through every type of filing, one can limit the filings displayed to a particular type by entering that type (e.g. N-CSR, N-30D, 485) in the "Filing Type" box in the upper left hand corner and doing a search by clicking the "Search" button centered near the top of the page.

    That returns a very similar looking "classic version" results page containing only the specified type of files, e.g. N-30D:

    image

  • edited July 2022
    It may be unclear what was meant by "backtracking". When one clicks on pages sequentially, there is a menu bar on the top that indicates path and one can backtrack to any page in the path. That was my "backtracking" reference to Company Search.

    I had no problems in reaching the indicated pages for VG 500, but my interest was in starting a new search.

    So, as I understand, @msf's steps, i) start search with the current Edgar, ii) switch to classic Edgar and further adjust search parameters (but don't start a new search), iii) get the desired archival/historical records. On hindsight, the "backtracking" I used just produced the current search page, not the archival/historical search page.

    Or2, find the CIK# for a new fund, and enter it in the URL of a previously saved archival/historical search results to start another archival/historical search. That also seems cumbersome.

    Or3, just start new search with this archival/historical search page as found by @mcq (I just couldn't reach this page directly from @msf's links).
    https://www.sec.gov/cgi-bin/srch-edgar

    Bottomline is that I have bookmarked both the current Edgar Search and Archival/historical Edgar Search pages for future use.
  • Here's a picture of the link to navigate (what Yogi calls backtracking) to the next level up:

    image

    For wonks: One can wind up at an unexpected "parent" page if one creates multiple paths (network organization) to a given page.

    See graph below. The blue lines present a "canonical" hierarchical structure. The red and green lines are alternative (network) paths.
    image

    Starting from the top, one may get to the third node at the bottom level two different ways:

    Follow the canonical hierarchy down: right, right, right, left; or
    Take an alternative path: right, right, left, right

    The latter follows a green edge (arrow) for the last step. Either way you get to the same place.

    When one backtracks, should it matter how one got to that node?

    There's no incorrect answer. If one says it should matter (path dependent) then one goes backward up the green arrow to the left. If one says it shouldn't matter (path independent), that the parent should be uniquely defined and one goes backward up the blue arrow to the right.

    Edgar's navigation takes the second, path independent approach. That's why "if one backtracks to general search page from either the new/current or Classic Edgar (and even the link provided by msf), one gets to the same search page/URL"

  • Please help a dummy. I cant’ understand what is the practical significance of this research. Please show me how one could use this data to compare the 5 year total return of two otherwise similar funds. Or in other words is this data point useful?
  • @larryB, current Edgar search is fine in most cases. But if you need to dig up older filings, then archival/historical Edgar search may work.
  • @yogibearbull. Thanks for your reply. BUT,,,, “current Edgar search is fine in most cases.” Fine for what? Fine for what practical use? Why would I need to dig up older filings? How would that data help me in 2022 find a meaningful difference in two funds?
  • edited July 2022
    ALL fund notices (changes, new/revised prospectuses, new funds) are filed at SEC/Edgar first. The info eventually is available on fund sites with the delays of up to a few weeks. @TheShadow does a great job of monitoring those in his posts. But Edgar is not a fund performance comparison site.
  • If you are concerned with performance over a brief interval like five years, especially the most recent five years, I can't imagine this discussion is relevant. Either EDGAR will give you the report(s) you need, and non-EDGAR sources will often be easier.

    My goal is different. I'm looking at historical performance over multiple decades, and more especially, year-by-year expense ratios. The pre-2000 reports, at least by Vanguard, contain year-by-year performance back 20 years or even to inception, which doesn't seem to be the norm for fund reporting today. Helpful when a fund has disappeared from M*

    Then as now the "Financial Highlights" section of the AR gives expense ratios for each of the trailing five years. So with mfs' help, I now have these back to 1989 (from 1994 reports), as opposed to being stuck at 1998, before I called for help. Prior to 1989 I go to my Wiesenberger yearbooks; and prior to 1945, to the 1939 SEC report.

    Yah, I don't get out much.
    larryB said:

    Please help a dummy. I cant’ understand what is the practical significance of this research. Please show me how one could use this data to compare the 5 year total return of two otherwise similar funds. Or in other words is this data point useful?

Sign In or Register to comment.