Author Topic: How Many Mutual Funds Routinely Beat The Market?  (Read 4761 times)

dungoofed

  • Pencil Stache
  • ****
  • Posts: 661
How Many Mutual Funds Routinely Beat The Market?
« on: March 15, 2015, 10:12:04 AM »
http://www.nytimes.com/2015/03/15/your-money/how-many-mutual-funds-routinely-rout-the-market-zero.html?_r=0

In other words, if all of the managers of the 2,862 funds hadn’t bothered to try to pick stocks at all — if they had merely flipped coins — they would, as a group, probably have produced better numbers. Instead of two funds at the end of five years, basic probability theory tells us there should have been three. (If you’re curious, I explained how the math works in a subsequent column, “Heads or Tails? Either Way, You Might Beat a Stock Picker.”

waltworks

  • Magnum Stache
  • ******
  • Posts: 3048
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #1 on: March 15, 2015, 10:29:59 AM »
Ouch.

-W

phillyvalue

  • 5 O'Clock Shadow
  • *
  • Posts: 88
  • Age: 27
  • Location: New York, NY
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #2 on: March 15, 2015, 10:36:07 AM »
The discussion of whether or not it's possible to beat the market has been rehashed too many times, on here and elsewhere. So let's not dig into that again.

But regardless, this is a very silly way to test whether or not active management can beat the market. It's crazy to think that anyone can beat the market over every successive 12-month period. This is short-term thinking at its worst. The only hope anyone has for beating the market in my view is investing for the long-term. Those who invest for the long term and ignore short term market fluctuations will undoubtedly underperform from time to time over short time spans.

That said, I am negative on most mutual funds for many reasons. It's a bad setup. Any "good" ones will get too big. Too much money flows in and out on a daily basis for the manager to focus on making attractive long-term investments. Moreover, few investors that are worth respecting will ever set up mutual funds, there is far more money to be made in alternative investments. Of all the mutual funds out there, there are only a handful I would ever invest in. The Sequoia fund is #1 on that list.

Look at their returns over time: http://www.sequoiafund.com/fp-investment-return-table.htm

You will see plenty of underperformance over that 45 year period. 

tj

  • Handlebar Stache
  • *****
  • Posts: 1236
  • Age: 33
  • Location: Maui
    • Arcadia Power
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #3 on: March 15, 2015, 12:04:34 PM »
IMO the goal should probably be more about keeping fees low and reasonable, avoiding loads. 12b1's etc, rather than completely avoiding active management altogether in order to save every penny in fund expenses.

http://m.kiplinger.com/article/investing/T041-C009-S002-primecap-best-stock-pickers-youve-never-seen.html

$10,000 invested in Vanguard Primecap Fund in 1984 would be worth over $500,000 today. The fund is now closed (as is the aforementioned Sequoia) and probably won't repeat at that scale, but if you invest in low cost actively managed funds, you're off to a better start than most people. Low cost could be firms like Wellington, PRIMECAP, and Dodge & Cox to name a few. Obviously Vanguard has several other active funds, but my problem with a lot of Vangaurd actively managed funds is that as the funds grow, they contract new advisors to manage part of the portfolio (such as the Windsor funds), rather than simply closing them like PRIMECAP.

Is it better than indexing? No idea. but there's some degree of comfort in paying negligibly more in fund expenses to have a finance professional analyze the balance sheets and pick holdings based on that rather than just buying everything that everyone is buying. I would rather use a low cost managed equity fund, such as Vanguard Equity Income Admiral Shares, for 20 bps, than pay a financial advisor 15-25 bps in addition to index fund expenses.

That being said, I will always recommend friends and family go directly to a Vanguard Target Date Fund.
« Last Edit: March 15, 2015, 12:07:30 PM by tj »

nereo

  • Walrus Stache
  • *******
  • Posts: 9657
  • Location: la belle province
    • Here's how you can support science today:
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #4 on: March 15, 2015, 12:34:00 PM »
But regardless, this is a very silly way to test whether or not active management can beat the market. It's crazy to think that anyone can beat the market over every successive 12-month period. This is short-term thinking at its worst. The only hope anyone has for beating the market in my view is investing for the long-term. Those who invest for the long term and ignore short term market fluctuations will undoubtedly underperform from time to time over short time spans.

... but this is what I always find so interesting.  On the very surface, my intuition says that someone who spends all their time researching companies *should* be able to beat the blindfolded monkey throwing darts.  How can knowing absolutely nothing beat knowing everything from free-cash flow to quarterly performance to economic trends? 
History of course has shown us that the blindfolded monkey wins, but it still surprises me.

What really shocks me here though is that, simply by random chance, there should have been 2 or 3 that finished in the top quartile five years in a row (random odds are 9.8 per 10,000).  But we get ZERO. Knowledge gets beaten by random luck.  J. H. Christ.

waltworks

  • Magnum Stache
  • ******
  • Posts: 3048
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #5 on: March 15, 2015, 01:02:59 PM »
Yeah, the thing that is interesting to me is not the absolute performance level, or whether the funds could stay at the top for X amount of time. The interesting thing is whether or not their performance was normally distributed - ie literally the same result (minus a little for some reason - not clear from the article if fees were included) as throwing darts. It seems likely that they are, which basically just means no matter what, you can't predict what the market or even individual stocks will do on a consistent basis, and anyone who does is just getting lucky.

I'm not quite that convinced that nobody is good enough to beat the market but man, there is no way in *hell* I would do anything but index based on the data.

-W

dungoofed

  • Pencil Stache
  • ****
  • Posts: 661
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #6 on: March 15, 2015, 01:20:12 PM »
I think the arbitrary time period skews the stats. But beware the active fund manager who uses it as a crutch - "oh yes, we'll beat the market in time," but (fees included) never does.

phillyvalue

  • 5 O'Clock Shadow
  • *
  • Posts: 88
  • Age: 27
  • Location: New York, NY
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #7 on: March 15, 2015, 01:21:10 PM »
But regardless, this is a very silly way to test whether or not active management can beat the market. It's crazy to think that anyone can beat the market over every successive 12-month period. This is short-term thinking at its worst. The only hope anyone has for beating the market in my view is investing for the long-term. Those who invest for the long term and ignore short term market fluctuations will undoubtedly underperform from time to time over short time spans.

... but this is what I always find so interesting.  On the very surface, my intuition says that someone who spends all their time researching companies *should* be able to beat the blindfolded monkey throwing darts.  How can knowing absolutely nothing beat knowing everything from free-cash flow to quarterly performance to economic trends? 
History of course has shown us that the blindfolded monkey wins, but it still surprises me.

What really shocks me here though is that, simply by random chance, there should have been 2 or 3 that finished in the top quartile five years in a row (random odds are 9.8 per 10,000).  But we get ZERO. Knowledge gets beaten by random luck.  J. H. Christ.

What you're saying and what I'm saying are two different concepts. There's the Ben Graham quote where he says that in the short-term, the market is a voting machine, and in the long-term, it is a weighing machine. In the short run, stock prices are a function of supply and demand, but in the long run, they will move in accordance with the value of the business. The consequence of this is that in order to deliver higher paper returns over every single short term period of time, you need to be more than a great investor. In fact, being a great investor and being able to identify undervalued companies would have nothing to do with it. You need to be a great psychologist who can consistently predict the behavior of other people. Believing that a manager can accurately predict the behavior of other people is, to me, a far crazier belief than thinking a manager can identify winning businesses that are undervalued.

A good investor can, will, and should underperform over certain periods, because in order to do something great, you have to be doing something other than the consensus. Buffett has underperformed at times, the Sequoia fund has underperformed over times, Baupost Group has underperformed at times, and yet over long periods, these folks have done very well.

nereo

  • Walrus Stache
  • *******
  • Posts: 9657
  • Location: la belle province
    • Here's how you can support science today:
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #8 on: March 15, 2015, 02:55:17 PM »
But regardless, this is a very silly way to test whether or not active management can beat the market. It's crazy to think that anyone can beat the market over every successive 12-month period. This is short-term thinking at its worst. The only hope anyone has for beating the market in my view is investing for the long-term. Those who invest for the long term and ignore short term market fluctuations will undoubtedly underperform from time to time over short time spans.

... but this is what I always find so interesting.  On the very surface, my intuition says that someone who spends all their time researching companies *should* be able to beat the blindfolded monkey throwing darts.  How can knowing absolutely nothing beat knowing everything from free-cash flow to quarterly performance to economic trends? 
History of course has shown us that the blindfolded monkey wins, but it still surprises me.

What really shocks me here though is that, simply by random chance, there should have been 2 or 3 that finished in the top quartile five years in a row (random odds are 9.8 per 10,000).  But we get ZERO. Knowledge gets beaten by random luck.  J. H. Christ.

What you're saying and what I'm saying are two different concepts. There's the Ben Graham quote where he says that in the short-term, the market is a voting machine, and in the long-term, it is a weighing machine. In the short run, stock prices are a function of supply and demand, but in the long run, they will move in accordance with the value of the business. The consequence of this is that in order to deliver higher paper returns over every single short term period of time, you need to be more than a great investor. In fact, being a great investor and being able to identify undervalued companies would have nothing to do with it. You need to be a great psychologist who can consistently predict the behavior of other people. Believing that a manager can accurately predict the behavior of other people is, to me, a far crazier belief than thinking a manager can identify winning businesses that are undervalued.
...
I understand what you are saying, and I generally agree with the Graham quote about the stock market being both a voting and a weighing machine (short v long term).  I also think 12 year periods are 'intermediate' time period - far longer than daily movements and including four quarterly reports, but still a far cry from the decades+ that most people will use for their retirement.  So... yes.  I would expect that someone with that much skill and experience to be able to be just a bit better than average every 12 months.  If not 1 year, what's the appropriate time-frame?  5 years?  10?  Sadly their records don't improve.
For me though what's really shocking is that not one of 2,862 could stay near the top, even though 3 should have been able to just with blind luck.

ChrisLansing

  • Bristles
  • ***
  • Posts: 341
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #9 on: March 15, 2015, 06:12:22 PM »
I think the arbitrary time period skews the stats. But beware the active fund manager who uses it as a crutch - "oh yes, we'll beat the market in time," but (fees included) never does.

Choose timelines carefully for best return :-)   

VTSAX has returned more than 20% over a 3 year time line.   5.78% since inception in Nov. 2000.   Since my time horizon is only 8 years I'm comfortable going with short term performance, which favors an index like VTSAX.   If my horizon was 30-40 years shouldn't I expect returns more in keeping with the longer time line? 


Indexer

  • Handlebar Stache
  • *****
  • Posts: 1293
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #10 on: March 15, 2015, 06:58:39 PM »
This isn't surprising at all. 

Previous studies came to the same conclusion. 

And for all the people complaining that they cherry picked the years. 

http://www.marketwatch.com/story/almost-no-one-can-beat-the-market-2013-10-25

"The study documents that in painful detail. Barras, Scaillet and Wermers tracked 2,076 actively managed U.S. domestic equity mutual funds between 1976 and 2006. They found that after fees, three-quarters of the funds exhibited zero “alpha,” a fund’s excess return over a benchmark index. And 24% of the funds were run by unskilled managers (who had negative alpha, or value subtraction).

And — are you sitting down? Only 0.6% — you read that right, 0.6% — showed any true skill at beating the market consistently, “statistically indistinguishable from zero,” the three researchers concluded."

Is 30 years a long enough time frame? 

hodedofome

  • Handlebar Stache
  • *****
  • Posts: 1211
  • Age: 39
  • Location: Texas
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #11 on: March 16, 2015, 08:39:58 AM »
Using mutual fund data to prove that the market can't be beat is like:

Using divorce court to prove marriage doesn't work

Like using a microscope to search and prove that there is no other intelligent life among the billions of other galaxies

Having a rocking chair. Gives you something to do but doesn't really get you anywhere.

nereo

  • Walrus Stache
  • *******
  • Posts: 9657
  • Location: la belle province
    • Here's how you can support science today:
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #12 on: March 16, 2015, 09:20:54 AM »
Using mutual fund data to prove that the market can't be beat is like:

Using divorce court to prove marriage doesn't work

Like using a microscope to search and prove that there is no other intelligent life among the billions of other galaxies

Having a rocking chair. Gives you something to do but doesn't really get you anywhere.
funny analogies.  I chuckled.  However, I don't think the authors of the study are claiming that the market cannot be beat, only that you can't beat the market by owning an mutual fund.  Two very different higns.

waltworks

  • Magnum Stache
  • ******
  • Posts: 3048
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #13 on: March 16, 2015, 09:28:44 AM »
Yeah, not great analogies. This is an honest attempt to look at the performance of a bunch of professional money managers. They generally sucked, and furthermore the ones that outperformed in any particular period failed to continue doing that in the future. That is bad if you want to believe you can beat the market by investing in actively managed mutual funds, and it's also bad for the general hypothesis that you have any decent chance at all of beating the market yourself. It is very, very strong evidence that persistent outperformance is the result of chance.

Obviously the situation could be different in other asset classes and fund types, the study didn't attempt to address that. But if there are NO fund managers in this study (out of a couple thousand) who could manage to consistently add any value, it's hard to argue that there are tons of people in other types of fund doing it.

-W

Using mutual fund data to prove that the market can't be beat is like:

Using divorce court to prove marriage doesn't work

Like using a microscope to search and prove that there is no other intelligent life among the billions of other galaxies

Having a rocking chair. Gives you something to do but doesn't really get you anywhere.
funny analogies.  I chuckled.  However, I don't think the authors of the study are claiming that the market cannot be beat, only that you can't beat the market by owning an mutual fund.  Two very different higns.

eyePod

  • Pencil Stache
  • ****
  • Posts: 966
    • Flipping A Dollar
Re: How Many Mutual Funds Routinely Beat The Market?
« Reply #14 on: March 16, 2015, 09:29:04 AM »
Random Walk Down Wallstreet does a really good job of explaining this. The best part? Even after they did a terrible job, you have to pay them more! Those trading fees are a bitch.