5 ms·

Misleading statistic: "We selected the 25 percent of funds with the best returns over those 12 months — and then asked how many of those funds actually remained

by cs702 12y ago

Misleading statistic: "We selected the 25 percent of funds with the best returns over those 12 months — and then asked how many of those funds actually remained in the top quarter in each of the four succeeding 12-month periods through March 2014." The statistic that matters, as Warren Buffett and others have repeated again and again, is COMPOUND RETURN!!! Consider these two investment options. Option 1 outperforms 60% of the time (three out of five years), but Option 2 produces a greater compound gain: YEAR Option 1 Option 2 Year 1 8.0% 15.0% Year 2 8.0% 7.0% Year 3 8.0% 4.0% Year 4 8.0% 19.0% Year 5 8.0% 6.0% Total gain 46.9% 61.4% Compound/yr 8.0% 10.1% The study I'd love to see is one that finds out (1) whether there are actively managed funds that consistently produce better compound returns than the indices over five to 10 year periods, and (2) whether there are any unusual concentrations of long-term compound outperformance that cannot be explained by chance among those funds. For example, what if it turns out that only a small number of actively managed funds are consistent long-term outperformers but many of them claim to use the same investment methodology? That would be a very unusual concentration of long-term success. I'd love to see a study with that kind of analysis.

liugiul 12y ago
What if you start investing in a bad year then?
- EliRivers 12y ago
  If you invested only at the peaks, right before crashes? Still works out great. http://awealthofcommonsense.com/worlds-worst-market-timer/ http://awealthofcommonsense.com/worlds-worst-market-timer/
  - dxhdr 12y ago
    http://www.acting-man.com/blog/media/2014/10/2-Nikkei-Long-term.gif http://www.acting-man.com/blog/media/2014/10/2-Nikkei-Long-t... Yes we're talking about the S&P500 but I'd like to point out that this isn't a general rule of markets; past performance does not guarantee future results. The S&P500 was at around 60 in 1960 and near 2000 today. It doesn't matter when you bought or at what intervals, money invested has performed well.
  - WalterBright 12y ago
    > If you invested only at the peaks, right before crashes? That's my patented investment strategy!
    - pbowyer 12y ago
      You're putting lots into the market this year then, I trust!
      - WalterBright 12y ago
        Absolutely! Chart any stock, and find it's high point. That's where I bought it.
        brianmcc 12y ago
        Please subscribe me to your "buying today" newsletter !
  - mortenjorck 12y ago
    Wow, that is a fantastic article and thought experiment. It certainly eased a bit of my own insecurity about investing "at the wrong time."
- walshemj 12y ago
  You are buying more for you money its called dollar/pound cost averaging
walshemj 12y ago
Quite you should be investing in a fund for a minimum of 5 to 10 years
guelo 12y ago
Wouldn't you just need to look at total returns? For example here's a list of mutual funds sorted by 5 year returns http://www.barchart.com/funds/5year.php http://www.barchart.com/funds/5year.php The S&P is up 77% over the last 5 years so all those funds beat the market. The problem is we don't know which funds are going to beat the market over the next 5 years.
- msmith 12y ago
  I'd just like to point out that the majority of those are sector funds, so it really doesn't make sense to compare it to a broadly-diversified index like the S&P 500. On top of that, many of the high-fliers in that list are leveraged ETFs, which magnify gains and losses, and would not be suitable at all for long-term portfolios.
  - IgorPartola 12y ago
    Why not? If I have $1 to invest, I can choose either and want to know which will give me a better return after X time. Should I go broad, or sector-specific? Which has more risk? Reward?
nostrademons 12y ago
It depends who you are and what you're looking for. If you're the fund manager, owner, or an individual investor, then compound return absolutely matters. You want the most amount of money at the end of the time period, and you've already committed your capital. But if you're a retail investor trying to decide where you want to put your capital - which, presumably, is who this article is aimed at - what you're looking for is whether the fund in question will outperform your alternatives. What this article is saying is "past performance is no guarantee of future results", i.e. you do just as well putting your money in an index fund or investing via a dartboard as you would picking the top-ranked fund in last year's returns. You don't get to travel back in time 5 years to make your investment decisions, so compound returns in the past are immaterial to you. You want the highest compound returns in the future, which on average come from passively investing in the indexes.
- tbrownaw 12y ago
  You don't get to travel back in time 5 years to make your investment decisions, so compound returns in the past are immaterial to you. You want the highest compound returns in the future, which on average come from passively investing in the indexes. 1. This is an article of faith. 2. "On average" doesn't matter. How much effort is required to stay away from the low end of that average is what matters.
- rbcgerard 12y ago
  The problem is that the article does not make that link. Because a manager does not remain in the top quartile of his/her PEERS does not mean they necessarily under performed the index...
a_c_s 12y ago
Great point! The only quibble I have is that 'compound return' doesn't include fees. However, 'total return' includes both fees & distributions (such as dividends). If we use fairly common fees, the table looks more like this: Option 1: 0.25% fee, Option 2: 1.5% fee Year 1 7.75% 13.5% Year 2 7.75% 5.5% Year 3 7.75% 2.5% Year 4 7.75% 17.5% Year 5 7.75% 4.5% Total gain: 45.2% 50.7% Compound/yr 7.75% 8.5% Once you include the fees (especially if the fund has a 'load' or sales fee) actively managed funds have to do significantly better than low fee index funds just to match total performance. Edit: formatting
- lucisferre 12y ago
  Why exactly do you think calculating compound return and fees are mutually exclusive? If you want compound return net of fees then you simply calculate that. As a rule most mutual funds do in fact report their performance net of fees.
nhaehnle 12y ago
Actually, I would argue that for the purpose of what the article was trying to convey, they used the right statistic. Let me illustrate by running with your own suggestion of trying to find out whether there are actively managed funds that consistently produce better compound returns than indices over 5 year periods. How would you actually do that? Certainly we agree that a fund outperforming an index over one 5 year period is irrelevant to answer that question. To check whether there is a fund that consistently outperforms an index over 5 year periods, you would look at several such periods, e.g. 5 consecutive 5 year periods (for a total of 25 years of data [1]). This is exactly analogous to the statistic used by the article, and somebody could then make the argument (like yourself) that a misleading statistic is used, and that people should instead look at the compound return over the entire period of 25 years. There is nothing wrong in principle with the statistic used by the article. However, it is reasonable to argue that looking at compound return over a 1 year period is too short (because typical investment time horizons are longer), and I would agree. Unfortunately, there's [1]. [1] And obviously, using such a long timespan is itself problematic for many reasons.
- rbcgerard 12y ago
  The problem is that article doesn't reference outperforming the index but rather being in the top quartile of peer performance...
  - nhaehnle 12y ago
    Right, that's absolutely a problem, but it's orthogonal to the supposed problem that OP was mentioning (e.g., it could be that those funds are so bad that even the top quartile doesn't outperform the index in a particular period). Or perhaps OP intended to refer to this problem, and it just wasn't done very clearly.
zsombor 12y ago
I would be surprised when a study comes out proving the opposite, given that there are strong proofs why beating the market average is difficult: http://www.norstad.org/finance/tsmproofs.pdf http://www.norstad.org/finance/tsmproofs.pdf. Sure none of these profs exclude the possibility of a small set of consistently winning super managers. But then again even if such superstars exist, ultimately their performance is bound to be self-limiting.
swatow 12y ago
nhaehnle's reply contains the fundamental error in your post, but there is another issue, which is that variance doesn't go away when you take the long run. Option 2 has more variance, and this doesn't get averaged out in the long run. Each year adds more variance to your wealth, which is a bad thing. Sure, the variance on the yearly rate of return will shrink as the time horizon grows. But because the yearly rate is compounded, this doesn't prevent the total variance in returns from growing with time. Long term investment doesn't cure the risk from volatile investments.

by cs702 12y ago

liugiul 12y ago

What if you start investing in a bad year then?

EliRivers 12y ago
If you invested only at the peaks, right before crashes? Still works out great. http://awealthofcommonsense.com/worlds-worst-market-timer/ http://awealthofcommonsense.com/worlds-worst-market-timer/
- dxhdr 12y ago
  http://www.acting-man.com/blog/media/2014/10/2-Nikkei-Long-term.gif http://www.acting-man.com/blog/media/2014/10/2-Nikkei-Long-t... Yes we're talking about the S&P500 but I'd like to point out that this isn't a general rule of markets; past performance does not guarantee future results. The S&P500 was at around 60 in 1960 and near 2000 today. It doesn't matter when you bought or at what intervals, money invested has performed well.
- WalterBright 12y ago
  > If you invested only at the peaks, right before crashes? That's my patented investment strategy!
  - pbowyer 12y ago
    You're putting lots into the market this year then, I trust!
    - WalterBright 12y ago
      Absolutely! Chart any stock, and find it's high point. That's where I bought it.
      - brianmcc 12y ago
        Please subscribe me to your "buying today" newsletter !
- mortenjorck 12y ago
  Wow, that is a fantastic article and thought experiment. It certainly eased a bit of my own insecurity about investing "at the wrong time."
walshemj 12y ago
You are buying more for you money its called dollar/pound cost averaging

walshemj 12y ago

Quite you should be investing in a fund for a minimum of 5 to 10 years

guelo 12y ago

Wouldn't you just need to look at total returns? For example here's a list of mutual funds sorted by 5 year returns http://www.barchart.com/funds/5year.php http://www.barchart.com/funds/5year.php The S&P is up 77% over the last 5 years so all those funds beat the market. The problem is we don't know which funds are going to beat the market over the next 5 years.

msmith 12y ago
I'd just like to point out that the majority of those are sector funds, so it really doesn't make sense to compare it to a broadly-diversified index like the S&P 500. On top of that, many of the high-fliers in that list are leveraged ETFs, which magnify gains and losses, and would not be suitable at all for long-term portfolios.
- IgorPartola 12y ago
  Why not? If I have $1 to invest, I can choose either and want to know which will give me a better return after X time. Should I go broad, or sector-specific? Which has more risk? Reward?

nostrademons 12y ago

It depends who you are and what you're looking for. If you're the fund manager, owner, or an individual investor, then compound return absolutely matters. You want the most amount of money at the end of the time period, and you've already committed your capital. But if you're a retail investor trying to decide where you want to put your capital - which, presumably, is who this article is aimed at - what you're looking for is whether the fund in question will outperform your alternatives. What this article is saying is "past performance is no guarantee of future results", i.e. you do just as well putting your money in an index fund or investing via a dartboard as you would picking the top-ranked fund in last year's returns. You don't get to travel back in time 5 years to make your investment decisions, so compound returns in the past are immaterial to you. You want the highest compound returns in the future, which on average come from passively investing in the indexes.

tbrownaw 12y ago
You don't get to travel back in time 5 years to make your investment decisions, so compound returns in the past are immaterial to you. You want the highest compound returns in the future, which on average come from passively investing in the indexes. 1. This is an article of faith. 2. "On average" doesn't matter. How much effort is required to stay away from the low end of that average is what matters.
rbcgerard 12y ago
The problem is that the article does not make that link. Because a manager does not remain in the top quartile of his/her PEERS does not mean they necessarily under performed the index...

a_c_s 12y ago

Great point! The only quibble I have is that 'compound return' doesn't include fees. However, 'total return' includes both fees & distributions (such as dividends). If we use fairly common fees, the table looks more like this: Option 1: 0.25% fee, Option 2: 1.5% fee Year 1 7.75% 13.5% Year 2 7.75% 5.5% Year 3 7.75% 2.5% Year 4 7.75% 17.5% Year 5 7.75% 4.5% Total gain: 45.2% 50.7% Compound/yr 7.75% 8.5% Once you include the fees (especially if the fund has a 'load' or sales fee) actively managed funds have to do significantly better than low fee index funds just to match total performance. Edit: formatting

lucisferre 12y ago
Why exactly do you think calculating compound return and fees are mutually exclusive? If you want compound return net of fees then you simply calculate that. As a rule most mutual funds do in fact report their performance net of fees.

nhaehnle 12y ago

Actually, I would argue that for the purpose of what the article was trying to convey, they used the right statistic. Let me illustrate by running with your own suggestion of trying to find out whether there are actively managed funds that consistently produce better compound returns than indices over 5 year periods. How would you actually do that? Certainly we agree that a fund outperforming an index over one 5 year period is irrelevant to answer that question. To check whether there is a fund that consistently outperforms an index over 5 year periods, you would look at several such periods, e.g. 5 consecutive 5 year periods (for a total of 25 years of data [1]). This is exactly analogous to the statistic used by the article, and somebody could then make the argument (like yourself) that a misleading statistic is used, and that people should instead look at the compound return over the entire period of 25 years. There is nothing wrong in principle with the statistic used by the article. However, it is reasonable to argue that looking at compound return over a 1 year period is too short (because typical investment time horizons are longer), and I would agree. Unfortunately, there's [1]. [1] And obviously, using such a long timespan is itself problematic for many reasons.

rbcgerard 12y ago
The problem is that article doesn't reference outperforming the index but rather being in the top quartile of peer performance...
- nhaehnle 12y ago
  Right, that's absolutely a problem, but it's orthogonal to the supposed problem that OP was mentioning (e.g., it could be that those funds are so bad that even the top quartile doesn't outperform the index in a particular period). Or perhaps OP intended to refer to this problem, and it just wasn't done very clearly.

zsombor 12y ago

I would be surprised when a study comes out proving the opposite, given that there are strong proofs why beating the market average is difficult: http://www.norstad.org/finance/tsmproofs.pdf http://www.norstad.org/finance/tsmproofs.pdf. Sure none of these profs exclude the possibility of a small set of consistently winning super managers. But then again even if such superstars exist, ultimately their performance is bound to be self-limiting.

swatow 12y ago

nhaehnle's reply contains the fundamental error in your post, but there is another issue, which is that variance doesn't go away when you take the long run. Option 2 has more variance, and this doesn't get averaged out in the long run. Each year adds more variance to your wealth, which is a bad thing. Sure, the variance on the yearly rate of return will shrink as the time horizon grows. But because the yearly rate is compounded, this doesn't prevent the total variance in returns from growing with time. Long term investment doesn't cure the risk from volatile investments.