Sunday, August 7, 2011

R.I.P. Active Fund Management

As I promised in my previous post, I've analyzed whether the average investor can use mutual funds to produce market beating returns. You have to understand that most individual investors will rely on the advice of an investment advisor, whether through a bank or insurance company, to purchase their retirement mutual funds. Both my experience and the evidence presented in the previous post show that investment advisors are good at 2 things: chasing past performance and failing to pick winning funds.

So if it were 2004 and I were listening to an advisor, s/he might suggest I invest in one of the top 10 funds in each asset class based on the last 3 years performance. This of course assumes this advisor has at their disposal the whole universe of Canadian mutual funds, which would be a very rare thing indeed. Most advisors, particularly those at banks, sell only their company's mutual funds. That goes for insurance companies as well. If you deal with an investment house like Edward Jones, they have a slightly broader offering, but not the whole universe. In fact, to access all the funds I researched, you'd have to be with a high price broker or use an online discount broker. You'll see why that would make no sense though because you'd be better off just using the very small and easy to understand universe of index funds.

Here's my methodology. I analyzed the rolling 3-year returns of mutual funds in the following 4 asset classes: Canadian equity, US equity, international equity, and Canadian fixed income. I used the data at FundLibrary for this purpose. I then identified the top 10 mutual funds in each rolling 3-year period. For the period starting 2004 (3-year average of 2003, 2002, 2001), I looked at the rank of the top 10 funds for the 10-year return ending 2010. For the top 10 funds in the 10-year return, I looked at where they ranked back in 2004. So I'm looking for 2 things here. If you invested in the top performers in 2004, did they end up performing big over 10 years? Also, looking at the top 10 funds based on 10 year performance, would there have been a way for you to easily identify them back in 2004?

The main conclusion from my research is that playing the mutual fund game is like rolling at the craps table. For example, of all the asset classes, the exact average ranking of the top 10 funds is 5.5. What is the average ranking in the 10-year performance measure? 16.33. The whole group as an average moves down 11 ranks. But when you look at the distribution of the rankings you see how much of a mess it is. While they start from 1-10 in 2004, they end up ranging anywhere from 1-72 on 10-year return. And it's no different when you look in reverse.

The top 10 funds based on 10-year performance, if past performance is a true predictor of future performance, should have been very easy to pick out among the universe of funds in 2004. While the top 10 over 10-years are by necessity ranked 1-10, they started out ranked anywhere from 1-88. Not quite as bad odds as the lottery but worse than some casino games.

But what if you just invested in the top funds based on highest 3-year performance? I did as such using GlobeInvestorGold. I made a hypothetical portfolio with equal weighting for each of the aforementioned asset classes. I invested $10 000 each year for 8 years. In the end, I had a personal rate of return of 1.71% per year, or a dollar growth of $5797.

I then created a similar portfolio but using only low cost TD e-series Index Funds mimicking the same asset classes. The result? 2.12% per year, or a dollar growth of $7275.

Now you might be saying, what's 0.5%? Well, over a 40-year investment span investing $10000 per year, it'll cost you about $53 000. And, the above projection significantly underestimates the underperformance of the actively managed approach for a few reasons.

1. Investment advisors often convince investors to sell and buy frequently in order to churn up commissions. This increases fees and reduces returns.
2. As documented in John Bogle's book mentioned in my previous post, most investors (and their advisors for that matter) don't have the discipline to stay on cruise control. Trying to time the market and chase performance always leads to diminished returns.
3. The projection above used No-Load funds when possible because the database seemed to have more data for them. Most of the time, if individuals are purchasing these funds through investment advisors, whether they know it or not, they are either paying a front-end load (pay a % fee when buying the fund) a back-end load (pay when you sell) or a deferred-sales charge. This would reduce returns further.
4. The top performing funds in many cases had fairly high initial investments so are unavailable to many individual investors.
5. I'm willing to bet MOST Canadians get their mutual funds through one of the big banks. There were a remarkably small amount of top performing mutual funds sold by the big banks. If I did a separate study showing just funds offered by those banks, it would not be pretty.
6. A good portion of the top performing funds were index funds. They actually skewed the performance higher. If I were to remove them, it would push the cumulative returns lower. I decided not to for 2 reasons. 1, I was tired of looking at data. 2, realistically, those funds are part of the pool from which an investor has to choose. Unlikely an advisor would recommend them, but you never know. Stranger things have happened.
7. Finally, survivor bias. By definition, all the funds I looked at still exist today. That means they are at least semi-respectable at performing or they would have been sacked. In John Bogle's similar study to mine, of 350 or so funds that started out the study, roughly 75% of them ceased to exist by the end of the study. But in hindsight, I can't include their performance because the data to which I have access no longer lists them. Their presence would significantly skew the returns downward. The other notable finding from my research is how few mutual funds on offer in each asset class even have 10 years of data under their belt. That's because the industry has to keep coming up with new and improved funds to draw in investors.

The final obvious conclusion from this little experiment is, even if you assume none of 1-7 to be true, and the performance difference is only 0.5%, the fact that there is a performance difference at all in favor of passive investing should put the whole argument to rest. And I repeat: I literally spend 5 minutes of my time each month managing my e-Series index portfolio. No meetings with investment advisors. No sales pitches for tied-in insurance products. I have asset diversification and low cost investing all in one package. Plus, every comparison I've ever run on my investments against those recommended to me by investment advisors have come out in my favor. (And yes, some of those comparisons have run from the time the investor made the recommendation, not looking in hindsight.) And no, it's not something of which only I'm capable. All you need is a little bit of knowledge, some patience, and some discipline when the market starts doing funky stuff.

As a final note, given how terrible the investment climate has been since I started investing in 2007, one would think my portfolio should be in the red. Well, it's nothing to write home about, but it is sitting at a respectable 2.8% per year. The total market has seen a decline over that same period. However, my superior performance is clearly due to dollar cost averaging. If you simply look at $10000 invested in November 2006 in index funds versus the theoretical market, the market is up 1.90% (this "market" being 25:25:25:25 as above) while the indexing is down 1.26%. How powerful then that simply investing a set amount of money at regular time intervals can increase my annualized return from a negative to a positive? Don't try and time the market. Let the ups and downs fall where they may.

There you go. Average is within your grasp. And you don't need a financial planner to get there. You may need one for other financial matters, but most certainly not for the majority of your retirement investing. And if and when you do need one, stick with a fee-only planner. You work hard for your money, so why should you be so quick to hand it over to someone else to take care of that knows little more about investing than what you can read in books?

Educate yourself. It's honestly not that hard. Just subscribe to MoneySense Magazine and you'll be set. Seriously. I've done a lot more reading than just that but for most people MoneySense will completely enlighten you and help you take control of your financial future.

No comments: