By Dan Hallett, CFA, CFP
Every time markets bleed red, mutual fund investors and the media become much more price-sensitive. So it’s no surprise that print media have featured a barrage of anti-fee articles. (See the National Post on December 6 and the Globe & Mail December 5 and December 9 articles.) Interestingly, Canada’s two national papers have united in a recent Investors Group bash-fest (see this Globe article, this National Post piece).
I agree with the broad message, which is to keep an eye on fees. Since fees are pretty transparent and easy to understand, they get most of the attention when trying to explain poor investor performance. But the media and the industry would do investors more good by identifying and trying to remove the handful of barriers to satisfying long-term performance.
Management Expense Ratio (MER)
Many of the articles casually mentioned a 2.50% MER figure as being average for Canadian mutual funds. Canadian mutual funds’ dollar-weighted MER at the end of 2009 was about 1.90% annually. Add in the effect of HST and we’re probably looking at fees of 2.1% per year. This doesn’t discredit any fee-driven arguments generally but it’s ideal to work with accurate figures.
Performance Dilution
Based on my observations from portfolio reviews of hundreds of investors do-it-yourselfers and advisors use too many funds when constructing portfolios. Finding a skilled manager is no cake-walk. But when you find one, you don’t want to dilute their potential to add value to your portfolio.
Portfolio managers with potential to outperform an index usually must be willing to invest differently than that index. But if you blend too many of these bold managers together into a portfolio, you run the risk of having a portfolio that is very index-like. This runs a higher likelihood of kicking out returns that more or less track an index with active management fees – which will doom most portfolios to long-term mediocrity.
Volatility Drag
You’ve no doubt scratched your head at why a portfolio’s long-term performance hasn’t quite lived up to expectations. It’s likely that volatility drag is one of the big culprits.
This is a very real factor explaining the gap between published performance and actual investor returns. The higher the volatility, the larger the volatility drag. And the larger the volatility drag, the wider the difference between published returns and investor performance.
If a mutual fund reports a 7 percent 10-year rate of return, for example, the only way to have achieved that precise result was to invest at the beginning of that period, hold for the full decade and have no buys or sells in between. But few people fall into that category.
The investment industry has long preached the benefits of investing a regular dollar amount so that you buy more units of a fund when the price goes down and fewer when it’s up. This intuitive argument just doesn’t hold.
Volatility actually works against you when buying or selling on a regular basis. Balanced fund investors might see performance that is 40 to 50 basis points annually less than published figures over a long period of time. Stock fund investors, however, might see returns that are 150 basis points (or 1.5 percentage points) less than published performance just from the fact that there are regular transactions over time.
In other words, periodic investing is a good idea because it’s a convenient way to save systematically. But make no mistake – it’s a more costly way to invest because the volatility usually reduces percentage returns of regular contributors.
Poor Timing
This cousin to volatility drag is a more frequently documented effect that can only be measured in hindsight. Whether you bought into funds after posting triple digit returns or sold out of a fund that sported deep double-digit losses, time usually provides you with the perspective needed to see where your timing went wrong.
Again, the higher the volatility the more likely it is that investors will be lured into an investment at the wrong time – and repelled by its poor performance just before a rebound.
The sum of poor timing and volatility drag explain the full difference between published returns and investor performance.
Taxes & Inflation
The impact of taxes varies widely from investor to investor. Mutual funds haven’t been the most tax efficient vehicles. However, the good news is that we have greater access to and make more use of tax-free and tax-deferred accounts. And more funds are designed tax-efficiently (though some are too costly). After-tax returns can be as much as 25% lower than net pre-tax returns.
And even low inflation rates like 2% per year can reduce real wealth by 18% over a decade and 33% over twenty years. So virtually any amount of inflation is more than material.
Improving Return Potential
While I’ve discussed the above factors in a mutual fund context, every one of them applies to investors buying stocks and most other securities. Quantifying and adding the above factors results in a frightening sum. They can easily add to a performance drag of at least 4% annually on a gross 10% total return. It’s worse for some investors.
The good news is that removing these performance hurdles – as much as possible – can go straight to investors’ bottom lines. Long-term performance can be enhanced by, for instance, keeping fees reasonably low, holding concentrated fund portfolios, minimizing investor mis-behaviour and using smart tax-cutting strategies. In other words, tilt the odds in your favour as much as possible.
You make some good points, although I’m not a fan of trying to find skilled managers; I’ve given up this pursuit and use indexing now.
I didn’t understand your final point about the effect of volatility on periodic purchases. Suppose we compare two scenarios where an investor makes monthly purchases of a fixed dollar amount. In the first secenario, asset pricess rise steadily with no volatility, and in the second scenario, asset prices rise to the same end point with volatility along the way. While it is possible for the second scenario to work out worse for the investor, the expectation is that the second scenario will work our better. Perhaps you were making a different comparison. Could you explain your thinking here?
Thanks for stopping by and posting a comment Michael.
I didn’t want to make this too long so didn’t go into detail but my comparison is based on actual data – not a scenario I cooked up. You can surely devise a scenario that would make either periodic investing or lump sum investing look better or worse.
But over a sufficiently long period of time, the volatility in financial markets – and the pattern of that volatility – has tended to play out the way I described above. Here’s an example.
When I look at all stock mutual funds reporting to IFIC (prior to this year when they stopped provide monthly data) I compared to investment scenarios. The first is a lump sum at the beginning of a 15 year period, holding for the entire time with no buys or sells in between. The second scenario features a fixed-dollar monthly investment at the beginning of each month (aka dollar cost averaging or DCA) .
The lump sum investment posted a percentage rate of return that was about 1.3% higher than the DCA scenario. If markets go into a long-term decline or if the front-end of the measurement period is ‘loaded’ with losses, the DCA scenario will likely look better.
When I tested this with lower volatility groups of funds, the gap was smaller. This simply confirmed a hypothesis I formed more than a decade ago, in the early years of my quantitative research on investor returns.
You can always find periods of time where this isn’t the case. But when you invest, the implicit assumption is that prices will rise – not fall – over time. Accordingly, unless there is nearly zero volatility, it’s very likely that making smaller more frequent amounts will result in a smaller percentage return compared to investing a lump sum and holding it for a very long time.
This breaks down, of course, where the holding period is short – in which case the ‘poor timing’ factor could rear its head.
As a supplement to my comments above, interesting investors and advisors should have a look at the slides from a recent talk delivered by my friend and former colleague Dr. Norm Rothery. The whole presentation is worth perusing but pay particular attention to slides 6 through 9, which relate to the above discussion.
I didn’t realize that you were comparing lump-sum investing to DCA. The main reason that lump-sum investing outperforms is because the average dollar is invested longer, not because of volatility. Perhaps you used some method to account for the differences in how long money was invested. I’d need to know how you did that compensation to understand the effect of volatility on your scenarios.
Effectively, yes because published performance assumes a lump-sump-type of scenario (i.e. buy at the beginning of the measurement period, hold through to the end, and have no interim trades). Periodic regular investing of a fixed-dollar amount is DCA. The factor you cite is a major factor affecting the total dollar amount accumulated at the end of the illustration but it’s a very minor factor in the calculation of a percentage rate of return – unless the total rate of return is very high.
I isolate this for any particular fund or group of funds by comparing the ‘real calculation’ with a a zero volatility scenario with the same time weighted return. The difference is the pure volatility drag.
Hi Dan,
Good piece overall and agree with the need for media & industry to focus on removing the barriers to long term performance.
At the same time, I’m not convinced the piece addresses the fee challenge put forth in the media. While you are indeed working with accurate figures, I’m not sure the spirit of quoting the dollar-weighted MER at the end of 2009 is fair. The reason this particular figure (1.90%) appears low is simply due to flight to safety/income and a shift in assets from higher MER long term/equity funds to lower MER money market or bond funds which occurred in 2009. Do you have any more recent MER data (i.e. 2011) to challenge the 2.50% claim?
I am a strong believer of indexing as the smarter investment approach to long term investing and will not recommend seeking skilled active managers. If we are advocating in favor of keeping an eye on fees, we need to help Canadians understand what they are paying for. I have not seen any data which shows that most active managers are able to beat the index consistently in the long term, hence difficult to justify the benefit of paying the higher cost associated with active management.
Cheers,
Silvio
Thanks for your comments Silvio. I’m not arguing with the media’s fee challenge as much as I’m trying to broaden their focus – mainly because there are other big threats to long term performance that are not that obvious and rarely written about. That’s the purpose of this piece.
As for the fairness of using 2009 data, I use it because it’s the most current data I have. I stand by the data because that figure was pretty consistent for several years. Back in 2003, Mark Warywoda (then of Morningstar) authored a great report on fund fees in Canada using 2002 data. His calculation for dollar-weighted MERs was 2.11% annually.
I calculated this same figure in 2007 and found the figure unchanged at 2.1%. Then followed that up in 2009 to see that it had fallen to 1.9%. It’s been a very stable figure for a long time so I’m very confident in simply taking the 2009 figure and adding the incremental HST impact. It’s certainly much more current than the standard 2.5% frequently mentioned or the now nine year old data used by a particularly flawed academic paper that received so much attention a few years ago.
But there is a reason why the “Fees” section was the smallest in my post above – because it’s not a major part of what I wanted to write about. Instead, I wanted to offer what I felt was a more accurate figure and move on to more important issues.
I respect your views with respect to the active vs. passive debate. I note, however, that indexing’s primary benefit is the cost advantage it offers to do-it-yourself investors. However, some balanced index funds are not materially cheaper than some of the great balanced funds I’ve been recommending for years – namely Mawer Canadian Balanced RSP and RBC Monthly Income (the latter available at fees below that of passive alternatives).
So, I think we need to stop thinking about it in terms of active vs. passive. Rather, this issue should be looked at in terms of what an investor needs in terms of exposure, what s/he has access to and, importantly, at what cost. That’s a more meaningful starting point than going straight into a general debate.
Thank you Dan. I will dig up some more recent MER data and share it.
Fully support your point with respect to the need for a broader focus and coaching investors to overcome other threats. We need to play an active role in this area and encourage the media to do the same.
Over the years there has been too much focus on the ‘product’ in the industry, pegging one fund against another, recommending great funds or the latest flavor. This often distorts the spirit of the clients’ best interests and does not support the broader focus. We need to forego the product debate and focus on the bigger picture.
Definitely agree with building a foundation based on what the investor needs in terms of exposure, creating a low cost & well diversified portfolio which fits their lifestyle and coaching them to stay the course when emotions & markets are @ extremes. That’s simple, wise and with an indexing strategy it can all be accomplished for all in cost of about 1%.
Cheers,
Silvio
It sounds like you are doing an IRR computation to compare DCA to a lump-sum investment. In this case, the DCA has the expectation of an edge because of volatility. Where DCA loses is when the better returns are in the early part of the time period. But volatility effects favour DCA.
Michael that’s the calculation methodology used to compare the scenarios but the outcome I calculated was not consistent with either your expectations, others’ expectations and the pitch put forth by the industry. You can certainly design scenarios or find periods of time where the outcome is as you describe, but that wasn’t the reality over the ~15 years of data I used.
I note that this method was validated in the mid-1990s in “Buy High, Sell Low: Timing Errors in Mutual Fund Allocations“, By Stephen Nesbitt, Fall 1995, The Journal of Portfolio Management. He calculated the timing errors of mutual fund purchases – i.e. comparing investor (dollar-weighted) returns to published (time-weighted) mutual fund returns. In part he wrote:
He later continued…
In his paper, he shows the published time-weighted returns of 17 fund categories, the dollar-weighted returns of the same categories and (as he described above) a dollar-weighted return based on a DCA scenario. Fourteen of the seventeen categories saw lower returns for DCA scenarios than for straight time-weighted (published) returns.
It’s certainly true that dollar-weighted returns are different from time-weighted returns. But this is due to performance chasing. I was looking at the narrow question of the effect of volatility because that’s what I thought you were discussing in your original piece. Once we add in other effects like poor choices by investors, I agree that returns will lag the index.
To be clear, I was looking at a comparison of IRR between 2 scenarios: 1) lump-sum invested in an index for a time period, and 2) DCA across the same period (with no discretion for the investor to time purchases or purchase anything other than the same index). For this question, the DCA approach would give a (very slightly) higher expected IRR due to volatility.
As is often the case when reasonable people disagree, it seems that we were actually answering different questions.
Michael, perhaps I wasn’t clear in my last note. Nesbit calculated all of it. My focus re: Volatility Drag is the equivalent of his: Time-Weighted Return vs. DCA Dollar weighted return. That is effectively the calculation I did for the Volatility Drag with minor differences as described further above. Note also that time weighted return equals dollar weighted returns for a lump sump investment.
It is this comparison that I relayed form Nesbitt’s paper – which effectively supports my volatility drag claim. In Nesbitt’s paper, DCA resulted in a lower IRR in 14 of 17 categories compared to a lump sum investment in those same categories.
I’ve explained this in as many ways as I can think of, so hoping that this clarifies the issue.
By the way, for those interested, Silvio sourced more up to date mutual fund MER at the end of 2010 and the asset weighted MER stood at 1.96% – still lower than my estimate. This may rise as the full effect of HST is captured in 2011 MERs. In 2011 we’ll see a full year of funds charging HST. Also, the 2010 figures are lower in part because many firms absorbed the extra tax until they decided how to deal with HST implementation. Those absorptions generally did not carry over far into 2011 – which is another factor that may nudge MERs up a bit.
I only have access to the first page of the Nesbitt paper, but it sets out to “measure the costs of market timing by analyzing the way investors’ movement of money between types of mutual funds reduces returns over time.” One way the describe this is the drag due to investors’ response to volatility. But it is the investors’ response to volatility that causes the drag, not the volatility itself. The disciplined investor who sticks to an asset allocation rather than just buying more of whatever performed best in the recent past will not be harmed by volatility.
Yes, Michael that was Nesbitt’s goal but his data table provides more of a data breakdown, including effectively the same figures used in my “volatility drag” estimate. Many papers contain interesting data that is not exlusively focused on the end goal or conclusion.
I cannot post more of Nesbitt’s article since it is copyrighted material but I’ve given you the information you need to look it up, perhaps at your local university library. If you do that, look up the table – hard to miss it’s a short paper with only one table.
I’ve done my best to address any potential misunderstanding. But there isn’t much I can do if you just don’t believe my conclusion. And that’s okay but there’s no point in continuing this back and forth in either case.
I have not done a good job of explaining what I’ve called Volatility Drag. So, that topic alone may warrant a follow up post so that I have the space to explain this more clearly…hopefully before the Christmas break.
[...] I agree that fees are extremely important, but there is more to the entire conversation than just that. Recently Dan Hallett, Director of Asset Management at HighView Financial Group added his two cents in a piece called Mutual fund critics missing the big picture. [...]
As promised, here is the follow-up to the above article – which focuses specifically on what I called Volatility Drag above. The new post more succinctly and accurately addresses the performance drag that DCA investors run into over the long-term.
Note that I’ve also dropped the Volatility Drag name in part because that term is used to describe the differences in arithmetic and geometric averages due to volatility. Also, there is more than volatility at work in causing a gap in percentage returns between DCA and lump-sump investing.
One last follow up on the fee issue, which wasn’t a big part of this article…While Silvio was kind enough to come up with a broader MER figure, Dave Chilton contacted me to say that the media’s figure of 2.5% is probably closer to the truth than not given that most are talking about equity funds.
My 2009 figure for equity fund MERs (asset weighted) was 2.1%. Adjusting for marginal amount of taxes (i.e. multiply by 113/105) brings the figure to 2.26% annually – which is close enough to 2.5% that I probably wouldn’t have mentioned it. Wanted to share this even though I was speaking about long-term funds generally.
[...] 3. Media misses the point…again! [...]
[...] use in general. According to Dan Hallett, Director of Asset Management at HighView Financial Group, the asset-weighted average equity fund MER in Canada was roughly 2.26% in 2009. However, as a counter point, a particular fund that has been dissected in the media lately was [...]