Wednesday, 4 May 2016

All investment returns are created equal, right?  Meaning for a series of valuations and cash flows there should be one unique solution.  Actually there are two methods to calculate the return of an investment portfolio - money-weighted and time-weighted.  Depending on the timing and size of cash flows both can provide noticeably different results.

Time-weighted investment returns can be thought of as the relative movement of your super fund investment option’s unit price. Money-weighted are the returns you’d experience on your individual portfolio.

Let’s define what these two different methods for calculating investment performance actually mean and their different methods for calculation.

Time-weighted return

When calculating a time-weighted return we’re attempting to remove the impact of money withdrawn or contributed to the portfolio.  It generally involves taking regular small sub-periods of revaluation (e.g. daily) and geometrically linking (compounding) those periods to generate a longer period return.  A time-weighted return does not factor in sequence risk, that is, the total investment return over the period will be the same regardless of the order in which the interim period returns occur.

It can be thought of as “what would the value of $1 invested at the start of the measurement period be worth now?"

It is this method of calculation which is generally communicated by asset managers for the following reasons:

  • it is simple to calculate
  • they feel it removes the impact of decisions out of their control, e.g. the timing of contributions and withdrawals into the portfolio
  • which in turn allows for easier direct comparison between different portfolios

The issue with a time-weighted return, from a super fund member perspective however, is that unless you are no longer contributing only a portion of your portfolio will ever actually experience the investment return that is reported.

Investment returns reported on superannuation fund websites or in research companies publications such as Chant West and SuperRatings, and even on member statements will invariably be time-weighted returns

Money-weighted return

On the other hand, the money-weighted return is a return calculation method which tries to capture the impact due to the size and timings of contributions and withdrawals in the portfolio.  It involves taking the opening and closing valuations and all the intermediate cash flows over a period and using an iterative numerical technique to solve for a return (or discount factor) that equates the present value of each cash flow (including the two valuation points) to zero. It is an equivalent measure to the yield to maturity on a bond. 

Money-weighted returns are important to consider because they better reflect the actual outcome a super fund member experiences.  This is because in accumulating a pool of capital to fund retirement the sequence of investment returns can have a significant impact on a final account balance and money-weighted return calculations try to reflect this.  A sequence of positive or negative returns on a larger account balance will have a larger impact on the final account balance but which would not be reflected in a time-weighted return.

What does all this mean in practice? Well from a member perspective the time-weighted method can also sometimes give counterintuitive results.  To illustrate this, consider the following scenario:

At the beginning of period 1, we contribute $10,000 – the return for period 1 is 25%

At the beginning of period 2, we contribute $15,000 – the return for period 2 is -10%

At the beginning of period 3, we contribute $25,000 – the return for period 3 is -10%

Under this situation the final balance is $44,775.

Relative to our contributions of $50,000 over the three periods we would say we had made a loss of $5,225.  Under a money-weighted methodology we would calculate an investment return of -6.4 per cent, consistent with our realised dollar outcome.  However under a time-weighted method we would calculate an investment return of +1.3 per cent.

So we can see that under a certain scenario the two different investment performance calculations can give significantly different results, and which one you prefer to use depends on your frame of reference.  Admittedly this is a technical topic to cover in just one post – if it’s of interest we’d be happy to explore this idea further in future posts.

The views of the author and those who provide the responses to comments posted on this blog are not necessarily the views of the QSuper Board. We’ve put this information together as general information only and you should get professional advice before relying on this information.

Past performance is not a reliable indicator of future performance. Each of our investment options has a different objective, risk profile, and asset allocation.