Time-Weighted Return – Only Tells Part of the Story
In my role as an Australian equity fund manager, the measure of investment performance that mattered to me – for both personal remuneration and for client assessment – was the time-weighted rate of return of the equity portfolio. This is the ‘track record’ that most fund managers proudly report in their product disclosure statements, their advertising material and industry performance surveys.
It is a well-established industry position, that when comparing the returns of different fund managers, where the objective is to compare the investment capabilities of those investment managers, the time-weighted rates of return should be used. For example, in APRA’s reporting standard on Investment Performance for MySuper products it defines the net investment historic return as “the time-weighted rate of return on investments, net of investment related fees, costs and taxes, adjusted for cash flows as they occur”. The CFA Institute’s Global Investment Performance Standards (GIPS) requires the use of time- weighted rates of return “because the performance measurement attempts to quantify the value added by investment decisions”.
Despite my acceptance that the time-weighted return is the best measure for comparing the historic results of investment managers in the same asset class over a specified period of time, I felt somewhat concerned that there was a disconnection between the reported time-weighted rate of return (normally represented by alpha generated against the benchmark) and the actual return achieved by the client. The reported time-weighted rate of return excludes the impact of cashflows, while the actual return achieved by the client was dependent on the timing of the cashflows.
Fund managers cannot be held accountable for the client’s timing of cashflows into their fund. That accountability should be with the client and forms the premise for excluding cashflows from return calculations when measuring performance of fund managers. This case, however, relies upon the assumption that the fund manager does not have any control of the timing of cashflows in or out of the fund.
It is also worth considering whether investment managers really have ‘no control’ over investment cashflows. Investment managers often have large marketing and business development functions aimed at attracting new clients and inflows from existing clients. Investment managers often also have product development functions. If they wish, investment managers are able to place capacity limits on funds and ‘close’ to new business.
Essentially, you are always a ‘bull’ (or at worst neutral) on your product when accepting inflows into your fund. If the portfolio has just outperformed the market by a large margin over the past few years, the marketing department, of course, sees this scenario as a ‘godsend’. It is time to ‘push’ this product in the market with a typical view of “let’s get out and promote the good performance”. There is no doubt that the inflows are highly correlated with past strong performance. However, a strong performance – particularly that achieved over a short period of time – is also often highly correlated with ‘expensive’ portfolios. This is typically seen in products exposed to ‘hot sectors’ of the market. It is also evident when buying demand, the inflows to a product pushes up the share prices of any relatively illiquid stocks in the portfolio.
Alternatively, if the portfolio has just underperformed the market by a large margin over the past few years, the marketing department will, of course, see this scenario as a ‘disaster’. A few of the existing clients might consider ‘topping-up’, but the marketing department would not even bother trying to attract new clients. You certainly wouldn’t advertise your poor past performance, in fact, you might even try to launch a new product to distract investors from prior products with the poor investment performance.
Through this process, a fund manager’s portfolio may outperform the benchmark when the portfolio is relatively small in size; and on the back of the past strong alpha generation, with product promotion, the fund may attract cash inflows. Once invested the fund may then underperform the benchmark. Overall, by using the time-weighted rate of return, the fund may still be able to show an investment rate of return above benchmark. This is despite the period of negative performance when the fund held the larger part, resulting in overall poor experience for clients.
The timing of cashflows may well often be outside his or her control, but I think there is a case for at least measuring the dollar value of alpha added by a fund manager. The dollar value of alpha added could be calculated based on the size of funds in a product and the level of alpha generated (excess return above benchmark). If a fund manager underperformed when the bulk of their money was under management, then this would well outweigh any outperformance prior to the bulk of the money flowing into the fund. In this case, the dollar value of alpha added would be negative. I believe measures that address the returns of the investor, rather than the fund manager, should become the primary focus of the industry.
I believe measures that address the returns of the investor, rather than the fund manager, should become the primary focus of the industry.
In addition to the time-weighted rate of return, the calculation of a dollar value alpha for an investment manager’s entire product suite would place the onus on the investment manager to only actively market a product when he or she considered the investment to be appropriate at that time. At the extreme, it may even encourage investment managers to return funds to clients when they consider the future investment returns from a product no longer appropriate for that client. It can be argued that the calculation of money-weighted returns or dollar value alpha would put back some onus on the investment manager to ensure products remain good investments for its clients. It may even discourage some of the more unsavoury industry practices, such as promoting ‘hot’ products at cycle peaks.
How well an investor times cashflows into and out of a fund will have a significant effect on that investor’s ultimate (money- weighted) investment return. I believe measures that address the returns of the investor, rather than the fund manager, should become the primary focus of the industry.
A positive example of this is that Platinum Asset Management’s Chief Executive Officer Kerr Neilson said in a recent investor roadshow presentation that of their long-term average reported return (time-weighted) of 13% p.a. compound, the average client return (money-weighted) was only about half of that. He said he would be delighted if his investors controlled their emotions and had some form of regular savings scheme.
In order to gain a better understanding in choosing the right solution to measure investment success, I would recommend reading the works of Yuri Shestopaloff and Alexander Shestopaloff – including A hierarchy of methods for calculating rates of return – 2007; Science of Inexact Mathematics, Investment Performance Measurement– 2009; and Solving the Puzzle of IRR Equation – 2011.
I hope this article encourages some industry debate. Please email any comments to the author.
CPD: Actuaries Institute Members can claim two CPD points for every hour of reading articles on Actuaries Digital.
