Putting Active Management to the Test

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
White Coat Investor
Posts: 13540
Joined: Fri Mar 02, 2007 9:11 pm
Location: Greatest Snow On Earth

Re: Putting Active Management to the Test

Post by White Coat Investor » Sun Oct 19, 2014 3:51 am

It's fun to see a true believer in active management. I didn't know there were any left. I figured all those active fund investors were just ignorant.

My question is how will the results of this change anyone's investing strategy. If 3/4s of his funds beat their benchmark, will someone go after active management? If only 1 out of 4 win, will he become an indexer?

And what happens if the funds disappear or the managers change? Does that give him an out?
1) Invest you must 2) Time is your friend 3) Impulse is your enemy | 4) Basic arithmetic works 5) Stick to simplicity 6) Stay the course

User avatar
packer16
Posts: 1044
Joined: Sat Jan 04, 2014 2:28 pm

Re: Putting Active Management to the Test

Post by packer16 » Sun Oct 19, 2014 6:44 am

I think a more interesting and related topic is what is active management? If the definition is narrowly actively managed funds then passive will be active primarily due to costs. However In fund studies, there has been outperformance by active managers before fees but the fess tip the scales to the passive guys. If the definition is more broad like investing different than the market benchmark, then active has and can beat passive investments via tilting and investing in more inefficient corners of the market like P-to-P lending or micro-cap (below $500m mkt cap)/illiquid to institutions or foreign partially open stock market stocks. One of the ways to reduce costs in these market benchmark beating strategies is to DIY versus hiring a manager.

Packer
Buy cheap and something good might happen

FinancialDave
Posts: 1585
Joined: Thu May 26, 2011 9:36 pm

Re: Putting Active Management to the Test

Post by FinancialDave » Mon Oct 20, 2014 11:14 am

EmergDoc wrote:It's fun to see a true believer in active management. I didn't know there were any left. I figured all those active fund investors were just ignorant.

My question is how will the results of this change anyone's investing strategy
. If 3/4s of his funds beat their benchmark, will someone go after active management? If only 1 out of 4 win, will he become an indexer?

And what happens if the funds disappear or the managers change? Does that give him an out?
I doubt that it will - either you believe the many books and studies that have already been done, or you want to try it out for yourself, OR you just leave all the investing to someone else, who really has no incentive to invest you in index funds.

The encouraging facts are that the totals in index funds are increasing. Of the $2 trillion held in the top 20 stock mutual funds, 23% is held in Vanguard's VTSMX & VFINX (TSM & SP500). The bad news is of these top 20 funds, 10 of them are loaded American Funds and over the last 3 & 5 years none of these American Funds have a better return than VTSMX. In fact only one fund even equals the VTSMX returns and that is the Dodge & Cox Stock Fund.


fd
I love simulated data. It turns the impossible into the possible!

barnaclebob
Posts: 3040
Joined: Thu Aug 09, 2012 10:54 am

Re: Putting Active Management to the Test

Post by barnaclebob » Mon Oct 20, 2014 11:33 am

DueDiligence wrote:
barnaclebob wrote:What is the average ER for these funds, Im guessing its close to .5 to .9%...
Good guess!
Overall average is 0.6%; many funds are close to this.
Vanguard average is 0.4% (investor shares), T Rowe Price average is 0.7%, Fidelity Low Priced FLPSX is 0.82%, Sequoia SEQUX is 1.0%.

Due Diligence
It wasn't a really a guess. When I owned a couple actively managed funds (inherited) they always underperformed within a few tenths of their ER. With this many active funds in the challenge we are looking at a psuedo index fund with a high ER. This set of funds is nearly guaranteed to under perform by their average ER in the long term.

User avatar
DueDiligence
Posts: 88
Joined: Mon Sep 09, 2013 3:57 pm

Re: Putting Active Management to the Test

Post by DueDiligence » Tue Oct 21, 2014 4:21 pm

FinancialDave wrote:
EmergDoc wrote:It's fun to see a true believer in active management. I didn't know there were any left. I figured all those active fund investors were just ignorant.
...either you believe the many books and studies that have already been done, or you want to try it out for yourself
I am unaware of work relevant to OP's question. Could either of you, or any other BHs, provide reference?

OP selected active funds with "a multi-tiered screening process for active funds...cost was the #1 most important factor, persistent risk-adjusted outperformance was important, portfolio turnover was significant, fund family history/reputation was a factor, and manager tenure was important but not as crucial as the other screening factors."

It appears that comparisons of active and passive do not attempt to screen-out active funds with known poor attributes such as high cost, poor performance that is persistent, high turnover, sold products etc. References that considered even one of the poor attributes, such as high cost (for example ER<0.5 or even only Vanguard active funds), would be of interest.

Respectfully

DueDiligence
"Personal preferences, circumstances, and abilities affect portfolio construction in a profound manner..." David Swensen

FinancialDave
Posts: 1585
Joined: Thu May 26, 2011 9:36 pm

Re: Putting Active Management to the Test

Post by FinancialDave » Tue Oct 21, 2014 5:18 pm

DueDiligence wrote:
FinancialDave wrote:
EmergDoc wrote:It's fun to see a true believer in active management. I didn't know there were any left. I figured all those active fund investors were just ignorant.
...either you believe the many books and studies that have already been done, or you want to try it out for yourself
I am unaware of work relevant to OP's question. Could either of you, or any other BHs, provide reference?

OP selected active funds with "a multi-tiered screening process for active funds...cost was the #1 most important factor, persistent risk-adjusted outperformance was important, portfolio turnover was significant, fund family history/reputation was a factor, and manager tenure was important but not as crucial as the other screening factors."

It appears that comparisons of active and passive do not attempt to screen-out active funds with known poor attributes such as high cost, poor performance that is persistent, high turnover, sold products etc. References that considered even one of the poor attributes, such as high cost (for example ER<0.5 or even only Vanguard active funds), would be of interest.

Respectfully

DueDiligence
I did not mean to imply that the OP's particular idea had already been tested, but that there have been publicized "challenges" in the past in the WSJ and other financial magazines pitting fund or stock picker's against the index, and as far as I know it has always ended badly for the "pickers." Heck I ran my own test of a mixture of 23 active and passive funds with my own money for 3 years, and there was nothing that lead me to keep any of the active funds after the 3 years, but I still have a select few index funds!

What you have to realize is that in your lifetime your couldn't prove that this method would work, you need to pick the funds and then let them sit for 30 years, unless you are going to play some kind of market timing game. Then when you are done you would need to prove that the results are statistically something more than luck. Then you would need to repeat the endeavor with another group for another 30 years.

I think it would be much easier to start by using history to scour the whole universe of mutual funds to find how many had beat their index for all 3 time periods of 10 years, 20 years, 30 years. Once you do this, I think you will find that the list is pretty small and there is not one set of criteria that would have allowed you to single out these funds 30 years ago.

Granted, all this says nothing about the future - just as if you were able to pick some funds that survived, that you would be able to do it again.

It hasn't stopped many from trying however.

fd
I love simulated data. It turns the impossible into the possible!

User avatar
Taylor Larimore
Advisory Board
Posts: 27512
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Re: Putting Active Management to the Test

Post by Taylor Larimore » Tue Oct 21, 2014 5:45 pm

It appears that comparisons of active and passive do not attempt to screen-out active funds with known poor attributes such as high cost, poor performance that is persistent, high turnover, sold products etc. References that considered even one of the poor attributes, such as high cost (for example ER<0.5 or even only Vanguard active funds), would be of interest.
DueDiligence:

Boglehead Rick Ferri screened-out 50% of the highest-cost actively managed funds and then did a comparison of similar index funds. The index funds outperformed. You can read about it on page 19 of his study.

The Case for Index Fund Portfolios

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

User avatar
DueDiligence
Posts: 88
Joined: Mon Sep 09, 2013 3:57 pm

Re: Putting Active Management to the Test

Post by DueDiligence » Fri Oct 24, 2014 4:01 pm

Taylor Larimore wrote: Boglehead Rick Ferri screened-out 50% of the highest-cost actively managed funds and then did a comparison of similar index funds. The index funds outperformed. You can read about it on page 19 of his study.

The Case for Index Fund Portfolios

Best wishes.
Taylor

Taylor kindly provided work by Ferri and Benke that compared 60:40 portfolios of active and investable index funds (not benchmarks). Ferri and Benke considered a number of scenarios including a sensitivity to ER where they filtered out active funds with ER greater than the median ER. Figure 1 summarizes their results for 3-fund portfolios.

Image
Figure 1: Ferri and Benke sensitivity to ER filtering, 3-fund 60:40 portfolios with 40% US equity and 20% International equity. The index funds are TBM VBMFX, TSM VTSMX, and TISM VGTSX. Rebalancing was not used. The median active portfolio underperforms by 1% in the base case and underperforms by about 0.5% in the ER filtered case (estimated from table, not provided by Ferri + Benke).

The ER filtering reduced, but did not eliminate, active underperformance; however the results seem encouraging IMO:
(1) Fraction of active portfolios underperforming was reduced from 80% to 70%.
(2) Underperformance of median active portfolio was reduced from 1% to 0.5%.
(3) Median outperformance of outperforming active portfolios was unchanged at 0.5%.
(4) Median underperformance of underperforming active portfolios was reduced from 1.2% to 0.9%.

Ferri and Benke suggested that additional filtering may be beneficial such as past performance, ratings, size, manager education, and fund age. It is not clear if additional filtering tests were made. Encouraged by their ER filtering, here are some results with extreme cost filtering using both/two Vanguard active 60:40 funds, Wellington VWELX and STAR VGSTX. Figures 2 and 3 summarize the results for VWELX and VGSTX respectively.


Image
Figure 2: Comparison of Wellington VWELX and two index benchmarks. The Ferri and Benke benchmark was weighted 60% VTSMX and 40% VBMFX with no TISM VGTSX since Wellington is mostly US. The Ferri-Benke benchmark does not include rebalancing. Results are also shown for VBINX Balanced Index since it does include rebalancing (Wellington rebalances). I believe that VBINX is the most defensible Ferri-Benke style benchmark for Wellington leading to Wellington outperformance estimate of 1.3% CAGR.


Image
Figure 3: Comparison of STAR VGSTX and two benchmarks. The Ferri and Benke benchmark is weighted 40% VTSMX, 40% VBMFX, and 20% VGTSX with no rebalancing. The STAR fund is active, rebalances, and had several changes in equity and fixed allocations over this time-period. Results are also shown for VSMGX LifeStrategy Moderate Growth since it rebalances and had similar changes in asset allocation to STAR. I believe that VSMGX is the most defensible benchmark for STAR because of rebalancing and allocations. STAR outperforms the Ferri-Benke style benchmark VSMGX by 1.1% CAGR.


The results show that both Wellington and STAR outperformed their 3-fund benchmarks by 1.3% and 1.1% annually for the 16 years 1997-2012. Outperformance of 1.3% is at the extreme outperformance tail of Ferri and Benke’s results shown above.

This does not prove anything or lead to conclusions or non-trivial observations. OTOH it is not satisfying (for me) to simply explain the results as randomness because of significant uncertainty and variability.

Perhaps Mr Ferri could comment as I may be misrepresenting something and he clearly understands this kind of work better than I.

Thoughts, comments, and recommendations are solicited and encouraged!

Respectfully

DueDiligence
Last edited by DueDiligence on Thu Oct 30, 2014 3:32 pm, edited 1 time in total.
"Personal preferences, circumstances, and abilities affect portfolio construction in a profound manner..." David Swensen

User avatar
DueDiligence
Posts: 88
Joined: Mon Sep 09, 2013 3:57 pm

Re: Putting Active Management to the Test

Post by DueDiligence » Thu Oct 30, 2014 3:30 pm

Here are results from applying the Ferri & Benke active vs passive methodology with filtering to all Vanguard active funds that existed in 1997. As noted earlier, Ferri & Benke constructed portfolios of 3 active funds by randomly selecting from all available active funds. They did a cost sensitivity by filtering out the 50% of funds with largest ER. They found that filtering reduced, but did not eliminate, the underperformance of active portfolios compared to the 3-fund portfolio.

I believe it is safe to say that no investor randomly selects from all active mutual funds. It seems more natural to filter first on the list of funds rather than selection methodology of random. For example, IMO randomly selecting from available funds at Vanguard or in an IRA or at a brokerage, is closer to investor practices than any method of selecting from all available funds. A negligible number of investors, retail or otherwise, have or want access to all available funds.

So I started with the Vanguard active, large-cap funds that existed in 1997. Allocation funds and sector funds (energy, GNMA, treasury, etc) were excluded and only intermediate taxable bond funds were used. The first figure shows the active funds consisted of 8 US large-cap equity, 2 intermediate bond, and 2 international equity. The funds were selected following Ferri and Benke methodology but restricted to Vanguard only. Ferri and Benke included all funds that existed over the period 1997-2012 and accounted for survivorship bias. I do not believe that any active Vanguard funds were terminated during this time. VBIIX Vanguard Intermediate Bond Index was classified as active since it is not cap-weighted (nominally 50% government and 50% corporate).

Image
Figure 1: CAGR(1997-2012) for Vanguard active funds conforming to Ferri and Benke selection process (large-cap, inception before 1997).

All combinations of the 8 US equity, 2 bond, and 2 international funds were considered for a total of 32 portfolios. Following Ferri and Benke, the portfolios were 40% US equity, 40% US bonds, 20% international equity, and not rebalanced. Figure 2 compares portfolio CAGR with the reference 3 fund portfolio, VTSMX+VBMFX+VGTSX.

Image
Figure 2: Difference of CAGR(1997-2012) between Vanguard active 3-fund portfolios and index 3-fund portfolio using Ferri and Benke procedures.

88% of Vanguard active portfolios outperformed and the median fund outperformed by 0.4% CAGR. All portfolios outperformed except for those that included US Growth VWUSX. Interesting to note that the best and worst portfolios contained a Vanguard large growth fund (Primecap VPMCX or US Growth VWSUX). Also interesting to note, as shown earlier, both Vanguard active 60:40 allocation funds Wellington VWELX and STAR VGSTX outperformed their benchmarks (not 3-fund portfolio) by 1.3% and 1.1% respectively.

These simple results are not comprehensive enough to arrive at general conclusions about active vs passive portfolios. However, IMO they are useful in conjunction with Ferri and Benke. The Ferri and Benke approach is realistic and valuable using portfolios, live/realistic benchmarks, and recognition of the need to filter active funds to replace unrealistic random sampling of all active funds. Their filtering based on ER is useful. A different first filtering could be on sold vs bought funds - high level differentiation of actual fund/investor types which showed significant impact in a recent paper, as I recall. Then filtering based on ER/cost would be interesting. The simple example of Vanguard accomplishes both types of filtering with extreme emphasis on cost.

Having said that, IMO this proves that one needs to be careful generalizing conclusions from studies based on the average of all available funds; not earth-shattering but useful to remember. It further demonstrates that additional filtering as suggested by Ferri and Benke might be useful and interesting.

Respectfully

DueDiligence
"Personal preferences, circumstances, and abilities affect portfolio construction in a profound manner..." David Swensen

User avatar
DueDiligence
Posts: 88
Joined: Mon Sep 09, 2013 3:57 pm

Re: Putting Active Management to the Test

Post by DueDiligence » Thu Oct 15, 2015 2:11 pm

Here is a 2 year update comparing performance of 24 active funds with their passive benchmarks. The 24 funds were selected based on ER, turnover, active strategy, past performance and other factors as discussed in OP. IMO the comparison is interesting since it represents how active funds might be selected. This is better than most active-passive comparisons that do not filter out funds with high ER, high turnover, bad strategies, poor past performance, or other criteria a real-world investor might employ.

The 1st figure shows the comparison for the year ending Sep 30, 2015. The 2nd figure shows composite results for 2 years from Sep 30, 2013.

Image
Figure 1: Summary active vs benchmark comparison for 12 months thru Sep 30, 2015.

Image
Figure 2: Summary active vs benchmark comparison for 24 months thru Sep 30, 2015.

The active funds averaged 0.3% higher returns for the year ending Sep 30, 2015 compared to 0.6% lower returns for the year ending Sep 30, 2014. The active funds had average composite underperformance of 0.1%/yr for the 2 years.

My takeaways based on the 2 year results:
(1) The difference between the active and passive funds is small and within the uncertainty (uncertainties include benchmark selection, fund class selection, and averaging method).
(2) The active Vanguard funds had relatively small deviations from their passive benchmarks, as observed in other threads. The T Rowe Price funds had larger deviations than the Vanguard funds. I don't know if this is generally true.
(3) To date the results support the comment of pkcrafter
pkcrafter wrote: I have also identified most Vanguard, TRPrice and Dodge and Cox funds as good active choices...As to their good performance, I just shrug and say so what?...If active funds underperform, the investor must make an uninformed decision of whether the underperformance is temporary or not...The confident investor may stay far too long, and the nervous one may switch much too soon, and both choices will hurt returns.
Paul


Comments appreciated.

DueDiligence
"Personal preferences, circumstances, and abilities affect portfolio construction in a profound manner..." David Swensen

User avatar
kenyan
Posts: 2930
Joined: Thu Jan 13, 2011 12:16 am

Re: Putting Active Management to the Test

Post by kenyan » Sun Nov 08, 2015 3:01 pm

Thanks for following up on this. Much more interesting to see the results of actual, documented predictions, fairly judged, than someone's backtesting or anecdotes.
Retirement investing is a marathon.

jay22
Posts: 705
Joined: Thu Aug 23, 2012 8:56 am
Location: Sacramento, CA

Re: Putting Active Management to the Test

Post by jay22 » Sat Nov 14, 2015 5:06 pm

This is a fascinating thread. I firmly believe that low cost, low turnover active funds can definitely boost the overall portfolio.

This year, we started a long-term position in POGRX and VGHCX with that in mind. These two funds now constitute 15% of our total portfolio.

User avatar
BackInTheBlack
Posts: 311
Joined: Tue Apr 27, 2010 4:37 pm

Re: Putting Active Management to the Test

Post by BackInTheBlack » Sat Apr 30, 2016 8:57 pm

I will post a much more detailed performance update when I have more time (very soon), but having just run the numbers tonight, it looks like, at first blush, 19 of 24 actively-managed funds from this list have outperformed their respective passively-managed benchmarks. That period of outperformance covers roughly the last 2.5 years (from early November 2013 to basically May 2016), and marks the quarter-way point in this 10-year thought experiment. I'm just posting this as a head's up, though, to look out for the forthcoming formal update.

While two and a half years is certainly not very long, 19/24, or 79.16% of funds outperforming, is still very significant. Is it difficult to pick actively-managed outperformers ahead of time? Absolutely. Is it impossible? The evidence would suggest not.
"Do not put your faith in what statistics say until you have carefully considered what they do not say." | | -William W. Watt

User avatar
BackInTheBlack
Posts: 311
Joined: Tue Apr 27, 2010 4:37 pm

Re: Putting Active Management to the Test

Post by BackInTheBlack » Sat Apr 30, 2016 9:06 pm

DueDiligence wrote:Here is a 2 year update comparing performance of 24 active funds with their passive benchmarks. The 24 funds were selected based on ER, turnover, active strategy, past performance and other factors as discussed in OP. IMO the comparison is interesting since it represents how active funds might be selected. This is better than most active-passive comparisons that do not filter out funds with high ER, high turnover, bad strategies, poor past performance, or other criteria a real-world investor might employ.

The 1st figure shows the comparison for the year ending Sep 30, 2015. The 2nd figure shows composite results for 2 years from Sep 30, 2013.

Image
Figure 1: Summary active vs benchmark comparison for 12 months thru Sep 30, 2015.

Image
Figure 2: Summary active vs benchmark comparison for 24 months thru Sep 30, 2015.

The active funds averaged 0.3% higher returns for the year ending Sep 30, 2015 compared to 0.6% lower returns for the year ending Sep 30, 2014. The active funds had average composite underperformance of 0.1%/yr for the 2 years.

My takeaways based on the 2 year results:
(1) The difference between the active and passive funds is small and within the uncertainty (uncertainties include benchmark selection, fund class selection, and averaging method).
(2) The active Vanguard funds had relatively small deviations from their passive benchmarks, as observed in other threads. The T Rowe Price funds had larger deviations than the Vanguard funds. I don't know if this is generally true.
(3) To date the results support the comment of pkcrafter
pkcrafter wrote: I have also identified most Vanguard, TRPrice and Dodge and Cox funds as good active choices...As to their good performance, I just shrug and say so what?...If active funds underperform, the investor must make an uninformed decision of whether the underperformance is temporary or not...The confident investor may stay far too long, and the nervous one may switch much too soon, and both choices will hurt returns.
Paul


Comments appreciated.

DueDiligence
To me, that quote from pkcrafter is probably the most instructive. Focusing purely on relative returns, while interesting, is also purely academic. It is far more telling to see what the related investor returns are, and when factoring in average investor behavior, passive management is definitely the better default investment. All I was attempting to prove in this experiment, however, was whether or not it is possible to consistently predict outperformance in such a way that can positively affect overall portfolio performance over time. It's still early in the life of this thing, so the verdict is undoubtedly still out, but this is still very much food for thought. Thank you duediligence for your helpful analysis. My apologies to the BogleHeads community for posting such occasional updates. After my more in-depth update next week, the subsequent should be at the halfway point, at the 5-year mark. In the inimitable words of John Bogle himself, "Press on regardless!"
"Do not put your faith in what statistics say until you have carefully considered what they do not say." | | -William W. Watt

User avatar
walletless
Posts: 837
Joined: Fri Aug 15, 2014 4:55 pm

Re: Putting Active Management to the Test

Post by walletless » Sat Apr 30, 2016 11:17 pm

Curious, what does your active portfolio or allocation look like?

User avatar
DueDiligence
Posts: 88
Joined: Mon Sep 09, 2013 3:57 pm

Re: Putting Active Management to the Test

Post by DueDiligence » Wed Oct 12, 2016 4:27 pm

Here is a 3 year update comparing performance of 24 active funds with their passive benchmarks. As described previously, the 24 funds were selected based on ER, turnover, active strategy, past performance and other factors. IMO the comparison is interesting since it represents how active funds might be selected. This seems better than most active-passive comparisons that do not filter out funds with high ER, high turnover, bad strategies, poor past performance, or other criteria an investor might employ.

Results for 3 years thru Sep 30, 2016.
Image
Image

The active funds averaged 1.5% lower returns for the year ending Sep 30, 2016 and 0.6% lower returns for the 3 years ending Sep 30, 2016. Underperformance of the active funds is almost entirely the result of a terrible year for Sequoia SEQUX, primarily a result of Sequoia's significant overweight in Valeant Pharmaceuticals.

Some observations:
(1) Vanguard and T Rowe Price were the only fund families with outperforming active funds.
(2) 6 of the 9 Vanguard funds outperformed their benchmark by an average of 0.4% per year.
(3) 6 of the 9 T Rowe Price funds outperformed their benchmark by an average of 0.2% per year.
(4) Both Dodge and Cox funds matched their benchmark.
(5) Active funds from the other fund families underperformed their benchmarks.

More details of the results can be found here .

The results continue to support the earlier comment of pkcrafter "I have also identified most Vanguard, TRPrice and Dodge and Cox funds as good active choices ...as to their good performance, I just shrug and say so what? ... If active funds underperform, the investor must make an uninformed decision of whether the underperformance is temporary or not ... the confident investor may stay far too long, and the nervous one may switch much too soon, and both choices will hurt returns"

Comments appreciated.

DueDiligence
"Personal preferences, circumstances, and abilities affect portfolio construction in a profound manner..." David Swensen

User avatar
neurosphere
Posts: 2980
Joined: Sun Jan 17, 2010 1:55 pm

Re: Putting Active Management to the Test

Post by neurosphere » Fri Oct 14, 2016 9:43 am

DueDiligence wrote: Underperformance of the active funds is almost entirely the result of a terrible year for Sequoia SEQUX, primarily a result of Sequoia's significant overweight in Valeant Pharmaceuticals.
And isn't that example one of the major problems with active funds? That in any one year any active fund may make a giant mistake from which it cannot recover? How does one mitigate against this risk? Perhaps by having a basket of various actively managed mutual funds for each asset class? The thought makes my head hurt. :D

User avatar
DueDiligence
Posts: 88
Joined: Mon Sep 09, 2013 3:57 pm

Re: Putting Active Management to the Test

Post by DueDiligence » Mon Oct 09, 2017 2:20 pm

Here is a 4 year update comparing performance of the 24 active funds with their passive benchmarks. The 24 funds had been selected in Sep 2013 based on ER, turnover, active strategy, past performance and other factors.
Difference between 4 year CAGR of benchmark and active fund for 4 years ending Sep 30, 2017.
Tick DiffCAGR4(Ben-F)
SEQUX 8.93%
MPGFX 2.25%
PRFDX 2.02%
OAKBX 1.68%
BERIX 1.47%
PRSIX 1.40%
FLPSX 1.23%
RPBAX 1.00%
VEVFX 0.81%
VWNFX 0.66%
VDIGX 0.65%
VEIPX 0.53%
TRMCX 0.02%
PRDGX -0.47%
DODBX -0.75%
VWELX -1.04%
DODFX -1.11%
OTCFX -1.49%
TRBCX -2.27%
VGENX -2.33%
VPMCX -2.39%
PRWCX -2.51%
RPMGX -2.80%
VHCOX -3.36%
Average 0.09%

-0.81% Average Vanguard(8)
-0.57% Average TRowePrice(9)
-0.93% Average Dodge&Cox(2)
3.11% Average Others(5)
13 Active Underperformed
11 Active Outperformed

The year ending Sep 30, 2017 was good for the active funds. At this time there is negligible difference between the average performance of the active and benchmark/passive funds.

The funds from Dodge & Cox, Vanguard, and T Rowe Price performed better than their benchmarks on average.

The 5 funds from the remaining fund families performed worse than their benchmarks by a significant amount, 5 of the 7 poorest performers were from the remaining 5 families.

The results continue to support the earlier comment of pkcrafter "I have also identified most Vanguard, TRPrice and Dodge and Cox funds as good active choices ...as to their good performance, I just shrug and say so what? ... If active funds underperform, the investor must make an uninformed decision of whether the underperformance is temporary or not ... the confident investor may stay far too long, and the nervous one may switch much too soon, and both choices will hurt returns"

One detail: TLTE substituted for EVAL in DODFX Benchmark#2 since Blackrock discontinued EVAL (EM value).

Comments appreciated.

DueDiligence
"Personal preferences, circumstances, and abilities affect portfolio construction in a profound manner..." David Swensen

User avatar
DueDiligence
Posts: 88
Joined: Mon Sep 09, 2013 3:57 pm

Re: Putting Active Management to the Test

Post by DueDiligence » Mon Oct 15, 2018 5:18 pm

Here is the 5 year update comparing performance of the 24 active funds with their passive benchmarks. The 24 funds had been selected in Sep 2013 based on ER, turnover, active strategy, past performance and other factors.
The 2018 results are only slightly different than 2017.

Difference between 5 year CAGR of passive benchmark and active fund, 5 years ending Sep 30, 2018.
Pas-Act(CAGR5)
VHCOX -3.3
TRBCX -2.9
RPMGX -2.4
VGENX -2.0
VPMCX -1.9
PRWCX -1.8
DODBX -0.7
VWELX -0.3
TRMCX -0.2
VWNFX -0.0
OTCFX +0.1
PRDGX +0.3
DODFX +0.4
VEIPX +0.4
FLPSX +0.5
BERIX +0.9
VEVFX +0.8
VDIGX +1.0
RPBAX +1.3
PRSIX +1.4
MPGFX +1.9
PRFDX +2.0
OAKBX +2.1
SEQUX +7.1

Averages
ALL24 +0.2
VAN_8 -0.7
TRP_9 -0.3
OTH_7 +1.7
10 Active Outperformed
14 Active Underperformed

The lowest cost/ER active funds (Vanguard) outperformed their benchmarks on average.
The intermediate cost/ER active funds (TRPrice) also outperformed on average.
The highest cost active funds (remaining fund families) underperformed on average.

DueDiligence
"Personal preferences, circumstances, and abilities affect portfolio construction in a profound manner..." David Swensen

User avatar
Taylor Larimore
Advisory Board
Posts: 27512
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Vanguard Total Stock Market Index Fund Admiral (VTSMX)

Post by Taylor Larimore » Mon Oct 15, 2018 6:30 pm

FinancialDave wrote: The bad news is of these top 20 funds, 10 of them are loaded American Funds and over the last 3 & 5 years none of these American Funds have a better return than VTSMX. In fact only one fund even equals the VTSMX returns and that is the Dodge & Cox Stock Fund.
Bogleheads:

VTSMX (Vanguard Total Stock Market Index Fund) is special. Not only has VTSMX beaten the American funds (assuming Dave's figures are correct), but it has also beaten its Morningstar category (large blend) average every year since its inception in 2001.

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

User avatar
DueDiligence
Posts: 88
Joined: Mon Sep 09, 2013 3:57 pm

Re: Vanguard Total Stock Market Index Fund Admiral (VTSMX)

Post by DueDiligence » Thu Oct 18, 2018 4:45 pm

Taylor Larimore wrote:
Mon Oct 15, 2018 6:30 pm
FinancialDave wrote: The bad news is of these top 20 funds, 10 of them are loaded American Funds and over the last 3 & 5 years none of these American Funds have a better return than VTSMX. In fact only one fund even equals the VTSMX returns and that is the Dodge & Cox Stock Fund.
Bogleheads:

VTSMX (Vanguard Total Stock Market Index Fund) is special. Not only has VTSMX beaten the American funds (assuming Dave's figures are correct), but it has also beaten its Morningstar category (large blend) average every year since its inception in 2001.

Best wishes.
Taylor
Taylor:
Dave's figures are misleading since (1) VTSMX is not a good benchmark for all the funds and (2) longer timeframes tell a different story.
The 9 largest American funds have outperformed their 3 fund benchmark by an average of 0.5% per year over the last 15 years.
Also these funds are now available without a load at most brokerages.

MStarTotalReturn%(10/17/2018) American-Benchmark
AFund Cat 5Year 10Year15Year 5Yr 10Yr 15Yr Comment
ABALX MA 08.26 10.67 07.48 +1.5 +1.6 +0.8 50%US,7%exUS,43%FI
AEPGX FL 03.80 08.27 07.44 +1.2 +1.1 +1.2
AGTHX LG 12.94 14.30 09.85 +1.2 +0.3 +0.6
AIVSX_ LB 10.95 12.46 08.50 -0.8 -1.6 -0.8
AMECX WA 6.55 9.88 07.37 -0.0 +0.7 +0.6 47%US,18%exUS,35%FI
ANCFX LB 11.44 13.57 10.11 -0.3 -0.4 +0.9
AWSHX LB 11.02 12.77 08.57 -0.7 -1.3 -0.7
CAIBX. WA 04.47 07.77 06.70 -1.2 -0.8 +0.1 36%US,34%exUS,30%FI
CWGIX. WB 06.49 09.91 08.54 +0.4 +0.3 +1.3 39%US,53%exUS,8%FI
VBMFX FI 01.73 03.69 03.70
VGTSX FL 02.59 07.18 06.21
VTSMX LB 11.73 14.02 09.26
Avg American(9)-Benchmark +0.1 -0.0 +0.5
Ref: Morningstar TotalRet%(10/17/2018)

DueDiligence
"Personal preferences, circumstances, and abilities affect portfolio construction in a profound manner..." David Swensen

Post Reply