New mutual fund reporting - art or science?
In doing my own research on mutual fund rates of return and even gain/loss calculations, I quickly found it to be a complex subject, even for someone that has a mathematical background. I consulted with widely respected analyst; Dan Hallett1, Morningstar Canada editor Rudy Luukko2 and Andre Fok Kam3 about some of the big changes regarding investors’ statements and reporting that we will see in 2017.
Mr. Luukko had some reservations about book values on client statements and cautions readers not to use them for their taxable gain/loss calculations. Book values (sometimes referred to as book costs) sometimes do not reflect their proper adjusted cost base used for tax calculations. Dan Hallett also questions the usefulness of book values. He says: “[Book values] may or may not reflect ACB or net amount invested. But for most, it’s unlikely to be an accurate measure of either.”
Here is another scenario where a book value may not reflect a true gain/loss. Because book values increase with each reinvested distribution, for some fixed income mutual funds that reinvest large distributions on a regular basis, you could theoretically encounter a situation where the book value exceeds the market value. Mathematically, this would show as a loss when in fact, the mutual fund could have a gain.
Why is ACB (Adjusted Cost base) so important? Tax professionals would use ACB to determine capital gain/loss for tax purposes. What about book values? Mr. Luukko says the intent of using book values was to comply with new regulatory requirements (CRM2) as securities dealers are mandated to provide gain/loss information to investors.
Dan Hallett has written an excellent article (with examples) on the new but very different way of calculating a rate of return called dollar or money weighted rate of return. A link is provided at the end of this article. Investors will be seeing these different performance numbers in a brand new Investor Performance Report that will be mailed out in 2017. In his article, Mr. Hallett cautions readers to be careful with these reported performance numbers, noting that calendar date biases may result in numbers that may not be meaningful in certain circumstances or comparing rates of return with other advisors may be difficult.
Securities regulators have decided that money weighted/dollar weighted return is to be used from now on. This formula is also referred to as the IRR (Internal Rate of Return). The return of the mutual fund itself uses a different formula called a time weighted rate of return.
Although the regulators have decided that advisors and their firms must use money weighted rate of return calculations, Mr. Hallett says he would have preferred to see time weighted calculations as did the CFA Institute. “Time weighted would have made so many things much easier – e.g., comparisons with broadly based indices; custom benchmarking; calculation of other performance metrics like Sharpe, alpha, etc.”
While regulators allow the advisor firms to use any method for internal reporting purposes, the money weighted return method must appear on the new Investor Performance Report that investors will be receiving in the mail in 2017. Some Dealers will be giving their advisors the option of adding time-weighted returns to advisor generated rate of return reports so investors can see both money weighted and time weighted rates of returns.
However, investors should be aware of some of the disadvantages of using money weighted (IRR) rate of return calculations. For instance, the formula presumes all cash flows during the time frame measured, are reinvested at the same internal rate of return – an assumption that some believe is completely unrealistic. Although empirically, the IRR is only the number or numbers that makes the Net Present Value of the cash flows = 0 in the IRR equation. If there are multiple rates of return that satisfies the equation, then that’s just the math. Having said that, in speaking with a mathematician - a Ph.D. in Applied Mathematics, there could also be a correlation between cash flow size and the possibility of multiple returns as well. He cautioned about “black box” formulas that do not handle IRR calculations very well and had some cautions about the use of Excel for IRR calculations.
Much of the modern literature seems focused on “fixing the IRR problem” by approximating the IRR formula using other equations. Ironically, the newly mandated use of IRR replaces the previous preferred rate of return model called the Modified Dietz Method which doesn’t have the multiple rate of return problem.
In the paper5 from Yuri Shestopaloff, Ph.D. and Wolfgang Marty, Ph.D. (Properties of IRR Equation with Regard to Ambiguity of Calculating of Rate of Return and a Maximum Number of Solutions) the authors appear to prefer the Modified Dietz Method for the calculation of rates of return.
If there are a number of swings of money coming in and going out of a mutual fund account, calculating an IRR rate may result in two or more multiple rates of return4. Or even worse –none. Mathematically speaking, each of these multiple returns are 100% correct and the observer is sometimes left with the difficult task of choosing one over the other. Sometimes it may not be intuitive which one is the “more correct” answer.
[Editor’s note: Although mathematicians can appreciate the beauty of pure math and report there can be more than one correct answer for the same polynomial equation (Descartes Rule), telling investors they could have multiple rates of return for the same mutual fund –all of which are perfectly correct –will make for some engaging discussions.]
Dan Hallett says that many clients have money moving in or out of their accounts and Descartes Rule can be a real issue for them.
Although going forward, calculating money weighted (IRR) rates of return (despite the challenges) could be for the most part, useful, there is a red flag regarding the transfer of mutual funds from one dealer to another. Mr. Hallett is referring to a dealer change when an advisor moves from one dealer to another. This is a very common scenario. In this case, the client’s mutual funds are transferred intact [this is called an in-kind transfer] from the relinquishing dealer to the new dealer at the market price the day the units are received by the new dealer. For the purpose of calculating a money-weighted rate of return (IRR), the original purchase cost(s) of the mutual fund do not move to the new dealer so the transfer-in of securities is posted at market value. If it is counted as a very large new purchase it could skew the IRR and it is debatable whether the rate of return would be meaningful. This will likely affect the “since inception” rate of return if you opened your existing account with an in-kind securities transfer. If the time frame measured excludes the securities transfer, all the other rates of return (except the “since inception” return) i.e. 1 year, 3 yr, 5yr or 10 yr returns should be OK.
Difficulties may also arise if a mutual fund is consolidated or merged with another mutual fund resulting in a large in-kind “purchase”. Again, this has the potential to skew returns. Your advisor is best equipped to discuss or alert you about these special situations.
However, Dan Hallett emphasized if the in-kind transfer is “new money” - say from another firm or another adviser, then it can be used as a basis for rate of return calculations.
Andre Fok Kam on the other hand, has a different viewpoint with respect to in-kind transfers of securities. According to Mr. Kam, an in-kind securities transfer is treated the same way as a new cash purchase of a mutual fund and under the letter of the law, the percentage rate of return generated would be perfectly valid under the new CRM2 securities rules.
Notwithstanding the potential issue of multiple rates of returns, if you never have an in-kind transfer in your account and opened your account with a cash purchase rather than a securities transfer, then all of your returns –including since inception should be OK.
Although convention would hold that the best way to measure performance is to compare original costs with a market value – transaction histories are not transferred when a client changes advisors or advisors changes firms. As a result, original costs are left behind. Although attempts have been made in the past to use book values(which incorporates original costs) for rate of return calculations, book values are incremented up every time you re-invest a distribution (so you aren’t taxed twice) in a taxable account. As a result, some say that book values should not be used for performance rate of return calculations at all.
Confusing? It can be. When you receive your new Investor Performance Report, bring it in to your Financial Advisor so both of you can review and discuss it in more detail.
1 Dan Hallett is Vice-President & Principal of Oakville, Ontario-based HighView Financial Group. Dan's views and opinions are regularly featured in the media and have appeared in several best-selling personal finance and investment books. He is a regular contributor to the Globe & Mail and many other business publications.
2 Rudy Luukko is editor, investment and personal finance, at Morningstar Canada. A former chair and founding member of the Canadian Investment Funds Standards Committee (CIFSC), he has also co-authored courses for the Canadian Securities Institute.
3André Fok Kam, CPA, CA, MBA is a consultant to the securities industry. He has advised regulators, fund managers, advisors and dealers.He has served on the Board of Directors of several fund managers, portfolio advisers and dealers and has served as Investment Director and CCO. He writes extensively for several trade publications.
4Descartes Rule: There is an indirect relationship between Descartes rule and the number of sign changes between positive and negative cash flows in or out of a portfolio ( money moving in and money moving out), then mathematically, there is a possibility of multiple rates of returns generated by the money weighted rate of return (IRR) formula. Search term: Descartes Rule of Signs.
5 Yuri Shestopaloff, Ph.D.,Wolfgang Marty, Ph.D. Properties of IRR Equation with Regard to Ambiguity of Calculating of Rate of Return and a Maximum Number of Solutions
Dan Hallett: Making sense of your new CRM2 performance report: http://www.theglobeandmail.com/globe-investor/funds-and-etfs/funds/making-sense-of-your-new-crm2-performance-report/article31876831/
Rudy Luukko: CRM2 won't solve tax-reporting headaches: http://cawidgets.morningstar.ca/ArticleTemplate/ArticleGL.aspx?culture=en-CA&id=692731