Choosing a KiwiSaver fund based on fee levels is analysis for suckers. We look at ten years of track record trying to find a relationship between fees and returns, unsuccessfully

The relationship between KiwiSaver fee levels and net returns is quite random

With the introduction of the KiwiFund bill to Parliament, that culminates a period of public discussion around the level of fees charged for KiwiSaver funds. It builds on a theme promoted by Sorted, and marketing by Simplicity.

The regulator, the FMA, also highlights the impact of fees on returns in its KiwiSaver Tracker tool.

And with ANZ and Kiwi Wealth cutting back fees on their default funds recently, that appears to justify the criticism.

There is no doubt that the international literature generally supports the notion that high fees inhibit the growth of members balances.

But most of that literature is based on the 10,000 fund-plus US market, the world’s deepest managed fund market and one that has been studied academically for decades.

But the New Zealand market is nothing like that. With less than 200 KiwiSaver funds, is it valid to use the US conclusions here?

Now that KiwiSaver has been going for ten years, we probably do have a track record to test.

As readers will know, we track KiwiSaver fund returns on a ‘regular savings’ basis and have done so since March 2008. It involves a life-cycle approach, mimicking the way a member has built their contribution, and how returns have built up over the business cycle as it has played out over the past ten years. It is quite a different approach to the fund industry’s point-to-point analysis (and one supported by the regulator, the FMA).

However, for this analysis, we have looked at what the fees were for each fund in 2017 on an equalised basis (that is, equalised with the same contribution level for all even though not all funds have been going since 2008). We have excluded funds who have not been going for at least three years, principally because we could not test our results on both a since-inception basis, and a last-three-years basis, which we like to do to be sure that results are not being skewed by some resting on earlier laurels.

First we looked at Default funds. The chart plots the 2017 fees (member scheme fees plus fund operating fees) and returns on an after-tax, after-all-fees basis.

This produces a somewhat surprising result. Rather than ‘proving’ that high fees result in low returns, it suggests quite strongly the opposite. For the technically minded, the statistical correlation of high-fees-give-high-results (low-fees-give-low-results) is a notable 0.7. But we don’t think you should jump to any conclusions about that; the sample is a tiny nine funds and both the returns and annual fee amounts are actually tightly bunched. (And when shifted to the last-three-years basis that correlation changes to 0.2.)

A better way to look at these Default funds is in the perspective of all KiwiSaver funds, like this:

The other risk categories reveal a wider and more random set of results.

For Conservative funds, you can pay over $600 in fees per year and get slightly better than average results (remember, these are ten year results on an average per annum basis). Or you can pay mid $300 per year and would have gotten outstanding results from one particular conservative fund, even better than any default fund and well worth the $100 or so that a default fund would have saved you in fees.

Among Conservative funds, there is a weak relationship between high fees and high returns.

For Moderate funds, it is clear that fee levels have little relationship to long term fund returns.

And that is equally true of Balanced Growth funds as well. Yes, you can choose a low-fee fund, but that hasn’t resulted in higher-than-average returns for this category. Equally, higher fees may or may not result in higher long-term returns. There is little correlation.

There is no practical correlation among Growth funds either.

And the category with the most funds, the Aggressive fund risk category, shows a particularly wide and random pattern. Choosing a fund based on fees here in the expectation that long term results will be improved would have been a very poor decision. And it would have been equally poor to assume that higher fees would have resulted in out-sized returns.

The bottom line is that you are wasting your time in the thought that fees are some sort of indicator of long term fund performance. For New Zealand KiwiSaver funds it is a false lead, a dry hole.

The link between fees and results is either just easy marketing to naive consumers, or superficial analysis by politicians.

If there is any even weak correlation, it is that low fees tend to result in low returns, but even that suggestion is not a slam dunk.

Public policy makers should be looking at other factors for outstanding fund management and results.

Professional fund managers know this. They are prepared to pay for enhanced long term results. But they don’t use the amount of the fees as the indicator.

You shouldn’t either.