You're missing the fact that your outcome depends on your decisions.TomCat96 wrote:Someone correct me if I'm mistaken. But I'm still of the opinion that volatility drag is a nonissue.
This is like one of those times someone goes into great elaboration of minor points, but fails to address the threshold question.
Why not simply look at the starting dollar value and the ending dollar value? That's all that matters.
....................
What am I missing?
Do you think about volatility drag on your portfolio?
Re: Do you think about volatility drag on your portfolio?
Last edited by *3!4!/5! on Fri Jun 16, 2017 3:14 pm, edited 1 time in total.

 Posts: 2306
 Joined: Fri Feb 23, 2007 8:21 pm
Re: Do you think about volatility drag on your portfolio?
TomCat,
Yes you are wrong. If two portfolios have the same simple average return, but one achieves that return with more volatility, then the final end value of the portfolio with less volatility will be greater: it will have had a greater geometric return. So the problem with your objection A is that the final value will differ depending on volatility.
Now we can only invest looking forward with expected returns and volatilities. But it is worthwhile to construct a portfolio aiming towards lower volatility for the same expected return.
Dave
Yes you are wrong. If two portfolios have the same simple average return, but one achieves that return with more volatility, then the final end value of the portfolio with less volatility will be greater: it will have had a greater geometric return. So the problem with your objection A is that the final value will differ depending on volatility.
Now we can only invest looking forward with expected returns and volatilities. But it is worthwhile to construct a portfolio aiming towards lower volatility for the same expected return.
Dave
Re: Do you think about volatility drag on your portfolio?
What part of what I wrote is wrong? This isn't me being antagonistic. I legitimately want to know which statement I wrote was erroneous.Random Walker wrote:TomCat,
Yes you are wrong. If two portfolios have the same simple average return, but one achieves that return with more volatility, then the final end value of the portfolio with less volatility will be greater: it will have had a greater geometric return. So the problem with your objection A is that the final value will differ depending on volatility.
Now we can only invest looking forward with expected returns and volatilities. But it is worthwhile to construct a portfolio aiming towards lower volatility for the same expected return.
Dave
edit:
It seems to me that I'm being misunderstood. I'm saying when I evaluate an investment, I'm going to look at the dollar amounts, the starting and ending dollar amounts.
My interpretation of your argument:
The final dollar amount is unknowable until you calculate it based on geometric returns
In other words, Tomcat96, you want to judge a portfolio based on its final dollar amount. But you can't do that. Instead you can only calculate the final dollar amount based on geometric returns. We judge portfolios based on expected returns.
And in light of the fact we judge portfolios based on expected returns, it is clear that a portfolio with lower volatility will provide gains superior to one with higher volatility.
When you respond to me, please say Yes or No, to whether or not I am understanding you correctly.
Last edited by TomCat96 on Fri Jun 16, 2017 4:13 pm, edited 5 times in total.
Re: Do you think about volatility drag on your portfolio?
I legitimately want to figure out what I'm doing wrong so I will try to present your arguments in the light most favorable to you.*3!4!/5! wrote:You're missing the fact that your outcome depends on your decisions.TomCat96 wrote:Someone correct me if I'm mistaken. But I'm still of the opinion that volatility drag is a nonissue.
This is like one of those times someone goes into great elaboration of minor points, but fails to address the threshold question.
Why not simply look at the starting dollar value and the ending dollar value? That's all that matters.
....................
What am I missing?
So I am Tomcat96 looking to make an investment decision. I don't know which portfolio to make. I see two prospectuses before me.
One says 9.9% returns, the other 10% returns with significantly greater volatility.
Is the worry that I will look at the 10% and think to myself that the 10% portfolio must be superior?
Objection A is that I will instead look at the starting and ending dollar amounts rather than any percentage gain, but I believe what you are saying is that I will not do that.
I believe what you are saying is instead I will only know the starting and ending dollar amounts from the annual gains.
Again what I'm writing is, let me look at what the starting and ending dollar amounts are. I don't care about what is reported for the annual percentage gains, instead tell me what the value of 10k invested in for ex. 1970 would be today.
What you are saying is that, I'm missing the fact that my outcome depends on my decisions.....
I'm missing something here.

 Posts: 2306
 Joined: Fri Feb 23, 2007 8:21 pm
Re: Do you think about volatility drag on your portfolio?
Tomcat,
I only read the top quarter of your post before I wrote my response. The initial assumption about initial and final value didn't make sense because final value will depend on volatilitynalong the way. I think you clarified that several lines down in your post, so I may have been overly harsh oops.
We design portfolios with an expected average return that would be the weighted average expected return of the portfolio components. The more uncorrelated, weakly correlated, negatively correlated components we can use to achieve the same mean return, the more likely we are to get the geo mean closer to the simple Mean. The problem is that more esoteric sources of uncorrelated return are more expensive to invest in. At what point is the improved portfolio efficiency outweighed by increased cost? Not at all sure! But I do think that the equations referenced above do indicate that cutting SD by 2% say from 16% to 14%, will increase Geo Mean by around 0.3%. Try the calculator at the bottom of the Canadian sister site page I referenced above.
I tried 3 portfolios, all with same 6% Mean return and different SD.
#1 Mean return 6%, SD 16%, GeoMean 4.8%
#2 Mean return 6%, SD 14%, GeoMean 5.1%
#3 Mean return 6%, SD 12%, GeoMean 5.3%
Dave
I only read the top quarter of your post before I wrote my response. The initial assumption about initial and final value didn't make sense because final value will depend on volatilitynalong the way. I think you clarified that several lines down in your post, so I may have been overly harsh oops.
We design portfolios with an expected average return that would be the weighted average expected return of the portfolio components. The more uncorrelated, weakly correlated, negatively correlated components we can use to achieve the same mean return, the more likely we are to get the geo mean closer to the simple Mean. The problem is that more esoteric sources of uncorrelated return are more expensive to invest in. At what point is the improved portfolio efficiency outweighed by increased cost? Not at all sure! But I do think that the equations referenced above do indicate that cutting SD by 2% say from 16% to 14%, will increase Geo Mean by around 0.3%. Try the calculator at the bottom of the Canadian sister site page I referenced above.
I tried 3 portfolios, all with same 6% Mean return and different SD.
#1 Mean return 6%, SD 16%, GeoMean 4.8%
#2 Mean return 6%, SD 14%, GeoMean 5.1%
#3 Mean return 6%, SD 12%, GeoMean 5.3%
Dave
Re: Do you think about volatility drag on your portfolio?
Random Walker wrote:Tomcat,
I only read the top quarter of your post before I wrote my response. The initial assumption about initial and final value didn't make sense because final value will depend on volatilitynalong the way. I think you clarified that several lines down in your post, so I may have been overly harsh oops.
We design portfolios with an expected average return that would be the weighted average expected return of the portfolio components. The more uncorrelated, weakly correlated, negatively correlated components we can use to achieve the same mean return, the more likely we are to get the geo mean closer to the simple Mean. The problem is that more esoteric sources of uncorrelated return are more expensive to invest in. At what point is the improved portfolio efficiency outweighed by increased cost? Not at all sure! But I do think that the equations referenced above do indicate that cutting SD by 2% say from 16% to 14%, will increase Geo Mean by around 0.3%. Try the calculator at the bottom of the Canadian sister site page I referenced above.
I tried 3 portfolios, all with same 6% Mean return and different SD.
#1 Mean return 6%, SD 16%, GeoMean 4.8%
#2 Mean return 6%, SD 14%, GeoMean 5.1%
#3 Mean return 6%, SD 12%, GeoMean 5.3%
Dave
This actually helps quite a bit. What are you saying then is
If we frame a problem in that we want to construct a portfolio which generates the best returns, then higher volatility will diminish those returns, and volatility drag is set of a principles and formulas which can actually compute by how much.
You are correct. I was looking at things ex post facto, because I didn't understand the question that volatility drag attempts to solve.

 Posts: 2306
 Joined: Fri Feb 23, 2007 8:21 pm
Re: Do you think about volatility drag on your portfolio?
Tomcat,
Exactly! We don't know what the actual returns and volatilities of various portfolio components will be in the future. But we can make modest reasonable assumptions about returns based on valuations and historical Premia for different factors. With regard to volatilities of different asset classes and correlations, history can be useful.
Dave
Exactly! We don't know what the actual returns and volatilities of various portfolio components will be in the future. But we can make modest reasonable assumptions about returns based on valuations and historical Premia for different factors. With regard to volatilities of different asset classes and correlations, history can be useful.
Dave
 Lieutenant.Columbo
 Posts: 1070
 Joined: Sat Sep 05, 2015 9:20 pm
 Location: Los Angeles CA
Re: Do you think about volatility drag on your portfolio?
Dave,Random Walker wrote:...The geometric mean (annualized) return of a portfolio is always less than the average annual return. This gap is caused by volatility, and the greater the volatility the greater the gap. An extreme example is gaining 100% one year and losing 50% the next year. The average return is 25%, but the annualized return (what really counts) is 0%.
Geometric return = Mean return  (0.5 X Variance)
Seems to me a strong reason to diversify across sources of return that are weakly correlated is to minimize the volatility drag. Of course this diversification comes at a cost. Is the cost worth it? I ultimately decided yes based more on faith than knowledge. Interested in what others think.
I just got here through your comment in the My Favorite Alternative Funds Topic and am going to read through it and will read Gibson shorter read you linked, but I wanted to go ahead and ask a basic (innocent) question, that might help me understand the need for Alternatives for the Larry Portfolio investor.
Apparently Small+Mid Value and highlyrated bonds have very low correlation during average market times and even Negative correlation during bear stock markets.
My question is: what would/could alternatives (like, say, using QSPIX/QSPRX for 10% of portfolio) add in terms of annualized return for the investor who is now, say, 50% in Global Small+Mid Value equities and 50% in highlyrated bonds?
Thank you.
Lt. Columbo: Well, what do you know. Here I am talking with some of the smartest people in the world, and I didn't even notice!

 Posts: 2306
 Joined: Fri Feb 23, 2007 8:21 pm
Re: Do you think about volatility drag on your portfolio?
LC,
I can't give you a great answer to your question: at least not a quantitative one. As I've said before, when evaluating a potential new addition to a portfolio what matters is expected return, volatility, correlations, and of course costs. What I can say, is that the initial most basic portfolio decisions are clearly the most important, have the largest marginal benefit, and are the cheapest. The biggest decision is the stock bond split. High quality bonds are the cheapest and best diversifier to a typical portfolio where 8590% of the risk is on the equity side. When one starts adding smaller doses of more expensive alternatives, the potential incremental benefit to the portfolio has to be smaller than the big equities / bonds decisions. But just because the marginal benefit is smaller does not mean it doesn't exist. If a small allocation nudges one's portfolio towards the northwest corner of some efficient frontier, then it can still be very worthwhile.
The alternatives have equity like after cost expected returns and no correlation to either equities or bonds. Whether one takes from the equity side, the bond side, or a bit from both, the portfolio's Sharpe ratio should be increased a bit. A lack of portfolio efficiency is a very real cost. The best estimates I've seen of this cost are expressed as the concept of volatility drag. I think decreasing a portfolio SD from say 16 to 12 is worth maybe 0.5% annualized. There are of course behavioral issues regarding volatility that can potentially be quantitatively much huger. From both a pure portfolio efficiency point of view and from a behavioral point of view, I think it makes tremendous sense to diversify across sources of return. The increased costs are certain and knowable, and the benefits only potential and not knowable. Each individual needs to do there best to recognize all the relevant issues and make their best decision.
Dave
I can't give you a great answer to your question: at least not a quantitative one. As I've said before, when evaluating a potential new addition to a portfolio what matters is expected return, volatility, correlations, and of course costs. What I can say, is that the initial most basic portfolio decisions are clearly the most important, have the largest marginal benefit, and are the cheapest. The biggest decision is the stock bond split. High quality bonds are the cheapest and best diversifier to a typical portfolio where 8590% of the risk is on the equity side. When one starts adding smaller doses of more expensive alternatives, the potential incremental benefit to the portfolio has to be smaller than the big equities / bonds decisions. But just because the marginal benefit is smaller does not mean it doesn't exist. If a small allocation nudges one's portfolio towards the northwest corner of some efficient frontier, then it can still be very worthwhile.
The alternatives have equity like after cost expected returns and no correlation to either equities or bonds. Whether one takes from the equity side, the bond side, or a bit from both, the portfolio's Sharpe ratio should be increased a bit. A lack of portfolio efficiency is a very real cost. The best estimates I've seen of this cost are expressed as the concept of volatility drag. I think decreasing a portfolio SD from say 16 to 12 is worth maybe 0.5% annualized. There are of course behavioral issues regarding volatility that can potentially be quantitatively much huger. From both a pure portfolio efficiency point of view and from a behavioral point of view, I think it makes tremendous sense to diversify across sources of return. The increased costs are certain and knowable, and the benefits only potential and not knowable. Each individual needs to do there best to recognize all the relevant issues and make their best decision.
Dave

 Posts: 868
 Joined: Sun Jul 09, 2017 2:53 pm
Re: Do you think about volatility drag on your portfolio?
Excellent answer. What alternatives are you utilizing at this point?Random Walker wrote: The alternatives have equity like after cost expected returns and no correlation to either equities or bonds. Whether one takes from the equity side, the bond side, or a bit from both, the portfolio's Sharpe ratio should be increased a bit. A lack of portfolio efficiency is a very real cost. The best estimates I've seen of this cost are expressed as the concept of volatility drag. I think decreasing a portfolio SD from say 16 to 12 is worth maybe 0.5% annualized. There are of course behavioral issues regarding volatility that can potentially be quantitatively much huger. From both a pure portfolio efficiency point of view and from a behavioral point of view, I think it makes tremendous sense to diversify across sources of return. The increased costs are certain and knowable, and the benefits only potential and not knowable. Each individual needs to do there best to recognize all the relevant issues and make their best decision.
Dave
 Lieutenant.Columbo
 Posts: 1070
 Joined: Sat Sep 05, 2015 9:20 pm
 Location: Los Angeles CA
Re: Do you think about volatility drag on your portfolio?
Dave,Random Walker wrote:I can't give you a great answer to your question: at least not a quantitative one. As I've said before, when evaluating a potential new addition to a portfolio what matters is expected return, volatility, correlations, and of course costs. What I can say, is that the initial most basic portfolio decisions are clearly the most important, have the largest marginal benefit, and are the cheapest. The biggest decision is the stock bond split. High quality bonds are the cheapest and best diversifier to a typical portfolio where 8590% of the risk is on the equity side. When one starts adding smaller doses of more expensive alternatives, the potential incremental benefit to the portfolio has to be smaller than the big equities / bonds decisions. But just because the marginal benefit is smaller does not mean it doesn't exist. If a small allocation nudges one's portfolio towards the northwest corner of some efficient frontier, then it can still be very worthwhile.
The alternatives have equity like after cost expected returns and no correlation to either equities or bonds. Whether one takes from the equity side, the bond side, or a bit from both, the portfolio's Sharpe ratio should be increased a bit. A lack of portfolio efficiency is a very real cost. The best estimates I've seen of this cost are expressed as the concept of volatility drag. I think decreasing a portfolio SD from say 16 to 12 is worth maybe 0.5% annualized. There are of course behavioral issues regarding volatility that can potentially be quantitatively much huger. From both a pure portfolio efficiency point of view and from a behavioral point of view, I think it makes tremendous sense to diversify across sources of return. The increased costs are certain and knowable, and the benefits only potential and not knowable. Each individual needs to do there best to recognize all the relevant issues and make their best decision.
I disagree, Dave: I think yours was a great answer. I wasn't expecting quantitative data, but the rational and the principles.
One annualizedreturn methodological/practical question now:
Say one wants to take, say, 5% from equities and 5% from bonds to allocate (a total of 10%) to QSPIX/QSPRX.
Say one has more than 10% room in Tax Advantaged (17% TaxDeferred, 3% TaxFree).
The annualized return resulting from adding the 10% QSPIX/QSPRX would be a little higher if the whole 10% is taken from TaxAdvantaged compared to taken from Taxable, correct? If you've addressed this elsewhere already, I apologize, as I must have missed it.
Thank you very much.
L.C.
Lt. Columbo: Well, what do you know. Here I am talking with some of the smartest people in the world, and I didn't even notice!

 Posts: 2306
 Joined: Fri Feb 23, 2007 8:21 pm
Re: Do you think about volatility drag on your portfolio?
LC,
I wrote that answer late last night and missed something obvious. I think one certainly can determine the expected return of a portfolio. Just take a weighted average of the expected return of the portfolio components. So if one takes from the bond side to generate alternative position, expected return increases, SD increases proportionately less, Sharpe ratio rises. If one takes from the equity side to generate the position, expected return same, SD decreases, Sharpe ratio rises. For two portfolios with same expected return, one with the high expected return asset classes being only equities and one equities and alternatives, the portfolio incorporating alternatives should have less volatility drag. Thus it's compounded return will be closer to the simple Mean return of portfolio components than the portfolio without alternatives.
If possible, by all means put the tax inefficient alternatives in tax advantaged space. But I believe there is one big exception. If one has municipal bonds in taxable and is willing to take on some increased risk because he is adding on totally different sources of return, then I think it makes sense to add the alternatives in taxable accounts. The after tax expected return of the alternatives I believe is less than equities, but greater than municipal bonds. I have LENDX, SRRIX, AVRPX in tax advantaged accounts and QSPRX, QMHRX in taxable accounts.
Dave.
I wrote that answer late last night and missed something obvious. I think one certainly can determine the expected return of a portfolio. Just take a weighted average of the expected return of the portfolio components. So if one takes from the bond side to generate alternative position, expected return increases, SD increases proportionately less, Sharpe ratio rises. If one takes from the equity side to generate the position, expected return same, SD decreases, Sharpe ratio rises. For two portfolios with same expected return, one with the high expected return asset classes being only equities and one equities and alternatives, the portfolio incorporating alternatives should have less volatility drag. Thus it's compounded return will be closer to the simple Mean return of portfolio components than the portfolio without alternatives.
If possible, by all means put the tax inefficient alternatives in tax advantaged space. But I believe there is one big exception. If one has municipal bonds in taxable and is willing to take on some increased risk because he is adding on totally different sources of return, then I think it makes sense to add the alternatives in taxable accounts. The after tax expected return of the alternatives I believe is less than equities, but greater than municipal bonds. I have LENDX, SRRIX, AVRPX in tax advantaged accounts and QSPRX, QMHRX in taxable accounts.
Dave.