Value Averaging

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Topic Author
mlewis
Posts: 209
Joined: Mon Nov 05, 2012 12:39 am
Contact:

Value Averaging

Post by mlewis »

A few questions on Value Averaging:

Do you think value averaging will increase returns or merely decrease risk? (although these are admittedly two sides of the same coin)

Would you value average your portfolio or the portfolio's individual components? I feel either method could pose its own theoretical problems.

Does long-term value averaging make sense for most people?
User avatar
LadyGeek
Site Admin
Posts: 71864
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Value Averaging

Post by LadyGeek »

For most people, the answer is no. Take a look at this wiki article: Value averaging

Here's why: In a down market, you'll need to have a large reserve available to make up the additional contributions. That can throw you for a loop and can be very difficult to manage.

We don't recommended value averaging for inexperienced investors.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
Xile F Investor
Posts: 135
Joined: Tue Jul 27, 2010 7:54 am
Location: Virginia

Re: Value Averaging

Post by Xile F Investor »

I read the wiki post.

I DCA into my Roth, but rebalance into the funds in the Roth (as part of my quarterly rebalancing). I contribute to my Roth once at the beginning of the year.

I thought I was doing DCA into IRA and Value Averaging into funds, but now it sounds like I'm not. How would one characterize what I'm doing? Is this a recommended or non-recommended approach?
What you must do is, speak so loud that I can not hear you. | Those who say it can not be done should not interrupt the people doing it.
User avatar
LadyGeek
Site Admin
Posts: 71864
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Value Averaging

Post by LadyGeek »

Here's the other half of the puzzle: Dollar cost averaging
Wiki wrote:Dollar cost averaging (DCA) [1] is the technique of dividing an available investment lump sum into equal parts, and then periodically investing each part.
If you are making regular, equal amount payments, it's Dollar Cost Averaging - a recommended approach. Where those payments go is a different question. Putting them into your Roth is a recommended way to do Rebalancing.

Value averaging is a different animal. It's not recommended unless you are prepared to have funds in reserve and the discipline to manage it.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
Random Walker
Posts: 4852
Joined: Fri Feb 23, 2007 8:21 pm

Re: Value Averaging

Post by Random Walker »

The expected return of the stock market is always positive; although sometimes the expected return is greater than others. So it makes sense to simply invest money as early as one can. Academic literature has shown lump sum investing immediately superior to DCA. Value averaging is really meant for a lump sum. For someone in accumulation phase I think putting money in as soon as available makes most sense. Now if one has a good diverse asset allocation with components that zig when others zag, if you use new additions to return towards your desired asset allocation, then you are adding a bit of value averaging flavor to DCA. I think of it as Dollar cost averaging on steroids :happy

Dave
Xile F Investor
Posts: 135
Joined: Tue Jul 27, 2010 7:54 am
Location: Virginia

Re: Value Averaging

Post by Xile F Investor »

@Dave
sounds like I'm doing what you described. Making 1/yr lumpsum contributions to IRA and allocating those contibutions to balance my AA.
I wanted to 1) figure out if I'm doing the right thing. 2) figure out what it's called.

@LadyGeek
you answered my question about if I'm doing VA, which I'm not. But since I don't hold funds in hand, I invest all, sounds like I'm not doing DCA either. Guess I'm doing Lump Sum annually w/ Rebalancing quarterly. Thanks.
What you must do is, speak so loud that I can not hear you. | Those who say it can not be done should not interrupt the people doing it.
User avatar
bertilak
Posts: 8374
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Value Averaging

Post by bertilak »

LadyGeek wrote:We don't recommended value averaging for inexperienced investors.
And we don't bother giving that advice to experienced investors because it is advice they don't need!

Value averaging leads to some tough questions. Here is how I evaluated value averaging when I was considering it as a strategy:

In order to value average you need to set aside some money (in cash or perhaps a low-volatility asset like short-term bonds). You therefore need to determine how much to set aside. Whatever you set aside will not be working for you as hard as it could be if it were in equities. Set aside too much and you are giving up up-side potential, too little and you don't have enough to effectively average-in during down markets. In other words you need an asset allocation. If you don't maintain that allocation you will be right back where you started and need to establish it all over again! That's rebalancing!

So, carrying value averaging to its logical conclusion eliminates the technique altogether.
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker (aka S.O.B.), the Cowboy Poet
User avatar
Topic Author
mlewis
Posts: 209
Joined: Mon Nov 05, 2012 12:39 am
Contact:

Re: Value Averaging

Post by mlewis »

Random Walker wrote:The expected return of the stock market is always positive; although sometimes the expected return is greater than others. So it makes sense to simply invest money as early as one can. Academic literature has shown lump sum investing immediately superior to DCA. Value averaging is really meant for a lump sum. For someone in accumulation phase I think putting money in as soon as available makes most sense. Now if one has a good diverse asset allocation with components that zig when others zag, if you use new additions to return towards your desired asset allocation, then you are adding a bit of value averaging flavor to DCA. I think of it as Dollar cost averaging on steroids :happy

Dave

Lump sum investing may beat out DCA/VA more often than not, but that does not necessarily make it the preferred strategy. Lump sum investing is also riskier, it increases the chances of a bad outcome or even e very bad outcome. If I suddenly received a large amount of money I would want to DCA or VA. Even if I am stil young and in the accumulation phase.

If you read Edleson's book, he talks about VA for long term accumulation, not for investing a single sum. However, I have more than once seen authors say things such as 'value averaging is a great way to deploy a large sum of money into the market over 1-3 years.' Perhaps they are right, and it is better for a large sum over a few years than it is for long term investing. I would certainly be more willing to do it with a lump sum.

As to your last comment, I don't understand why using additions to return to your desired asset allocations means adding a value averaging flavor. This is just rebalancing. And no matter what your zig/zag looks like, the components will not stay on a value path.

thanks
malcolm
User avatar
bertilak
Posts: 8374
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Value Averaging

Post by bertilak »

mlewis wrote:If I suddenly received a large amount of money I would want to DCA or VA.
So, once fully invested at whatever asset allocation you are aiming at, why don't you sell everything (get that large amount of money back again) and start over -- again and again and again ... ?

If you don't think selling everything once you are fully invested is a good idea, what is it a good idea to be in the same position at the beginning? Once fully invested, whatever chances there are of sudden market drop are exactly the same as they were on day one. If you were afraid at day one, there is no reason to be less afraid once you are fully invested.
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker (aka S.O.B.), the Cowboy Poet
User avatar
Topic Author
mlewis
Posts: 209
Joined: Mon Nov 05, 2012 12:39 am
Contact:

Re: Value Averaging

Post by mlewis »

bertilak wrote: So, carrying value averaging to its logical conclusion eliminates the technique altogether.
Bertilak,

I've had similar thoughts as you describe with relation to VA. However, I don't agree with your conclusion. Value averaging does supposedly increase the IRR, even if your final outcome is not a greater amount of money. The same return for less risk seems like a good deal to me. If you decrease risk through VA, you might be able to increase risk in another way, by increasing your stock allocation, for instance.
User avatar
Topic Author
mlewis
Posts: 209
Joined: Mon Nov 05, 2012 12:39 am
Contact:

Re: Value Averaging

Post by mlewis »

bertilak wrote: So, once fully invested at whatever asset allocation you are aiming at, why don't you sell everything (get that large amount of money back again) and start over -- again and again and again ... ?

If you don't think selling everything once you are fully invested is a good idea, what is it a good idea to be in the same position at the beginning? Once fully invested, whatever chances there are of sudden market drop are exactly the same as they were on day one. If you were afraid at day one, there is no reason to be less afraid once you are fully invested.
Perhaps, but by using DCA/VA you are not simply hedging against the idea that you might lose money. You are hedging against the idea that you may have bought in at a time when things were overvalued. You don't want to buy in at a market peak. No matter what the long term market averages tell you, if you buy when things are overvalued your long term returns will be lower. So, if you DCA/VA at the outset, you are giving a little security to your long term expected returns.
User avatar
Topic Author
mlewis
Posts: 209
Joined: Mon Nov 05, 2012 12:39 am
Contact:

Re: Value Averaging

Post by mlewis »

Here is an old article on DCA people might like.

http://www.efficientfrontier.com/ef/997/dca.htm

malcolm
Random Walker
Posts: 4852
Joined: Fri Feb 23, 2007 8:21 pm

Re: Value Averaging

Post by Random Walker »

M Lewis,
I agree, what I am describing is rebalancing with new money. The point of value averaging is to buy more when prices are down and less when prices are up. Rebalancing with new money does that as well. But it is buying low at the asset class level rather than the portfolio level. I also agree with what you said about lump sum versus DCA. But more often than not, over the long term lump sum is superior.
The most important thing is to have a good AA that you are willing to stick to with new additions. That is what I mean by DCA with value averaging flavor. It has the added advantage of investing the money sooner, and the expected return of the market is always positive. But if you want to value average that is certainly fine too. As you know William Bernstein is a big fan. If you are into building financial muscles and rigorous zen like financial discipline (and I'm all for that!), value averaging will accomplish that big time.
One big advantage of value averaging that I see is that it makes the investor focus on a specific ultimate accumulation goal rather than just "save a lot and make it grow". Curious to hear what you think.
Lastly, I think specifics such as your age and the size of the lump sum relative to your income/regular savings are very significant in the personal value averaging decision.

Dave
User avatar
LadyGeek
Site Admin
Posts: 71864
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Value Averaging

Post by LadyGeek »

FYI - We have an 11+ page thread for discussing the differences between lump sum and dollar cost averaging: Finally! Can we lay DCA to rest and ban this topic forever? (feel free to continue the debate...)

This thread is between value averaging and lump sum, which I believe is a different perspective.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
Topic Author
mlewis
Posts: 209
Joined: Mon Nov 05, 2012 12:39 am
Contact:

Re: Value Averaging

Post by mlewis »

Random Walker wrote:M Lewis,
I agree, what I am describing is rebalancing with new money. The point of value averaging is to buy more when prices are down and less when prices are up. Rebalancing with new money does that as well. But it is buying low at the asset class level rather than the portfolio level.
This gets back to one of my original questions. If you VA the portfolio as a whole and rebalance back to your desired AA of your components, then are your portfolio components really on a VA path? And does it matter? On the other hand, if you VA the individual components, then your AA will get out of whack since the components will have different expected returns. This certainly can matter, especially over the long term. Can you combine VA of the components with rebalancing? Would that make sense? This is what I'm most unsure about.

I also agree with what you said about lump sum versus DCA. But more often than not, over the long term lump sum is superior. True, but you achieve that greater chance of superiority at the expense of greater risk, and the greater probability of a very bad outcome. Take your risk where you like it I suppose.
The most important thing is to have a good AA that you are willing to stick to with new additions. That is what I mean by DCA with value averaging flavor. It has the added advantage of investing the money sooner, and the expected return of the market is always positive. But if you want to value average that is certainly fine too. As you know William Bernstein is a big fan. If you are into building financial muscles and rigorous zen like financial discipline (and I'm all for that!), value averaging will accomplish that big time.
One big advantage of value averaging that I see is that it makes the investor focus on a specific ultimate accumulation goal rather than just "save a lot and make it grow". Curious to hear what you think. Perhaps. But if you have the discipline to VA, you probably have the discipline to focus on a goal anyway. Whether you VA over the short or long term, you face the possibility of not meeting the goal, so that could also lead to disappointment?
Lastly, I think specifics such as your age and the size of the lump sum relative to your income/regular savings are very significantly in the personal value averaging decision.
I agree. Being young, I would be more likely to just put it all in at once. However, if I received a large amount of money that significants increased my net worth or was much larger than my normal savings rate, I would probably be inclined to VA, no matter how old I was.

Dave
Malcolm
Random Walker
Posts: 4852
Joined: Fri Feb 23, 2007 8:21 pm

Re: Value Averaging

Post by Random Walker »

Malcolm,
If you value average at the portfolio level, I think that implicitly assumes the new additions are added so that you rebalance towards your designated AA. So with value averaging I would view it as buying low/selling high at both the asset class level and the portfolio level. I imagine that can only benefit the overall portfolio. You are definitely getting at the limits of my knowledge. I read the book and definitely did not understand all of it.

Dave
rr2
Posts: 1041
Joined: Wed Nov 19, 2008 10:04 pm

Re: Value Averaging

Post by rr2 »

mlewis wrote:A few questions on Value Averaging:

Do you think value averaging will increase returns or merely decrease risk? (although these are admittedly two sides of the same coin)

Would you value average your portfolio or the portfolio's individual components? I feel either method could pose its own theoretical problems.

Does long-term value averaging make sense for most people?
Search for older threads, especially posters tpm871, avalpert. Both have been missing from this forum recently. See also this thread: http://www.bogleheads.org/forum/viewtopic.php?t=75863
User avatar
bertilak
Posts: 8374
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Value Averaging

Post by bertilak »

mlewis wrote:]Value averaging does supposedly increase the IRR, even if your final outcome is not a greater amount of money. The same return for less risk seems like a good deal to me. If you decrease risk through VA, you might be able to increase risk in another way, by increasing your stock allocation, for instance.
DCA does decrease risk, but it does so by reducing (temporarily) your equity allocation. This also reduces your up-side potential to the degree you are below your target equity allocation. That is no mystery. With DCA you will eventually reach your target allocation and therefore your target exposure to equities and the associated risk. If that risk was not acceptable at time X1 what makes it acceptable at time X2?

VA can increase or decrease your risk depending on if it is currently indicating more or less exposure to equities. Your up-side potential is proportionally increased or decreased at the same time. Again no mystery. Contrary to what you say above you do NOT have the same expected return while your equity exposure is reduced nor do you have the same risk when your equity exposure is increased.

The point of VA is to time the market so you increase your equity exposure at low points in the market and decrease it at high points. This is a form of market timing. I do believe some people can be somewhat successful in market timing, but those who are successful will base their timing on market fundamentals, not on what your personal asset allocation happens to be at any given time. The market does not know or care about your plans.
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker (aka S.O.B.), the Cowboy Poet
User avatar
Topic Author
mlewis
Posts: 209
Joined: Mon Nov 05, 2012 12:39 am
Contact:

Re: Value Averaging

Post by mlewis »

bertilak wrote:
DCA does decrease risk, but it does so by reducing (temporarily) your equity allocation. This also reduces your up-side potential to the degree you are below your target equity allocation. That is no mystery. With DCA you will eventually reach your target allocation and therefore your target exposure to equities and the associated risk. If that risk was not acceptable at time X1 what makes it acceptable at time X2?

There are different types of risk. If you DCA a sum of money over a period of time vs lump sum, you are still shooting for the same AA once you are invested and the risks that such exposure will entail, but you are reducing the risk that you could be investing all of your money at a particularly unfortunate time of over-valuation and low ERP.

VA can increase or decrease your risk depending on if it is currently indicating more or less exposure to equities. Your up-side potential is proportionally increased or decreased at the same time. Again no mystery. Contrary to what you say above you do NOT have the same expected return while your equity exposure is reduced nor do you have the same risk when your equity exposure is increased.

In a sense I agree. But the whole idea of VA is based around expected returns. As the market rises and falls the valuations that affect the long term ERP also change. This affects your projected outcome/dispersion of outcomes from that point forward. I don't believe that the risk of owning equities changes much, but the ERP does change. If you accept that idea, then the risk/return profile of investing in stocks is not always the same. VA seeks to invest less when the ERP is lower and more when it is higher. I have since returned the book to the library, but I do remember Edelson's data showing rather conclusively that the IRR of the VA portfolios bested the DCA portfolios in his long term investment programs. To me, that says he is getting more return for less risk, or less risk for the same return, or however you like it.

The point of VA is to time the market so you increase your equity exposure at low points in the market and decrease it at high points. This is a form of market timing. I do believe some people can be somewhat successful in market timing, but those who are successful will base their timing on market fundamentals, not on what your personal asset allocation happens to be at any given time. The market does not know or care about your plans.

The process of VA does base it on fundamentals, which are based around your initial expected return, which also should have been based on fundamentals. Every type of investing is some form of market timing, unless you never put money in the market. You just have to decide the best way to put it in that you feel maximizes your return for the risk you are willing to take.
Malcolm
umfundi
Posts: 3361
Joined: Tue Jun 07, 2011 5:26 pm

Re: Value Averaging

Post by umfundi »

mlewis wrote:A few questions on Value Averaging:

Do you think value averaging will increase returns or merely decrease risk? (although these are admittedly two sides of the same coin)

No. No.

Would you value average your portfolio or the portfolio's individual components? I feel either method could pose its own theoretical problems.

No.

Does long-term value averaging make sense for most people?

No. Only for those who can predict the market. Which is no people.
Keith
Déjà Vu is not a prediction
Bill Bernstein
Posts: 604
Joined: Sat Jun 23, 2007 12:47 am

Re: Value Averaging

Post by Bill Bernstein »

Good thread.

One small point: DCA + rebalancing is almost mathematically identical to value averaging.

Bill
rr2
Posts: 1041
Joined: Wed Nov 19, 2008 10:04 pm

Re: Value Averaging

Post by rr2 »

wbern wrote:Good thread.

One small point: DCA + rebalancing is almost mathematically identical to value averaging.

Bill
Can you expand how this is?
User avatar
Topic Author
mlewis
Posts: 209
Joined: Mon Nov 05, 2012 12:39 am
Contact:

Re: Value Averaging

Post by mlewis »

wbern wrote:Good thread.

One small point: DCA + rebalancing is almost mathematically identical to value averaging.

Bill
I could be wrong, but this would seem less true in more extreme conditions.
If you had a very high equity exposure, for instance, in the event of a big worldwide bear market, your overall portfolio value would be very different with DCA + rebalance compared to VA.

malcolm
umfundi
Posts: 3361
Joined: Tue Jun 07, 2011 5:26 pm

Re: Value Averaging

Post by umfundi »

wbern wrote:Good thread.

One small point: DCA + rebalancing is almost mathematically identical to value averaging.

Bill
Possibly. Both of which are inferior to Lump Sum Investing. Even before you say, + Rebalancing.

Is there really a question here?

Keith
Déjà Vu is not a prediction
Bill Bernstein
Posts: 604
Joined: Sat Jun 23, 2007 12:47 am

Re: Value Averaging

Post by Bill Bernstein »

Note I qualified with "almost."

If you have a portfolio that's very large relative to your inflows, *and* if you've got a very high equity exposure, *and* if there's a humongous decline, then, yes, DCA + RB will tell you to buy less stock than stand-alone VA.


But, as a practical matter, in that very unusual circumstance, with VA you almost certainly will not have enough cash to buy as much equity as the discipline tells you to, so it comes out pretty much the same in the end as with DCA + RB.

Bill
User avatar
Topic Author
mlewis
Posts: 209
Joined: Mon Nov 05, 2012 12:39 am
Contact:

Re: Value Averaging

Post by mlewis »

umfundi wrote:
wbern wrote:Good thread.

One small point: DCA + rebalancing is almost mathematically identical to value averaging.

Bill
Possibly. Both of which are inferior to Lump Sum Investing. Even before you say, + Rebalancing.

Is there really a question here?

Keith
Most people are accumulators, (or spenders) and don't have the option to lump sum invest most of the time.

malcolm
User avatar
Topic Author
mlewis
Posts: 209
Joined: Mon Nov 05, 2012 12:39 am
Contact:

Re: Value Averaging

Post by mlewis »

umfundi wrote:
mlewis wrote:A few questions on Value Averaging:

Do you think value averaging will increase returns or merely decrease risk? (although these are admittedly two sides of the same coin)

No. No.

Would you value average your portfolio or the portfolio's individual components? I feel either method could pose its own theoretical problems.

No.

Does long-term value averaging make sense for most people?

No. Only for those who can predict the market. Which is no people.
Keith
Keith,
Again, just to clarify, my initial questions were not really with comparison to lump sum investing. I was asking them with respect to using VA as a long term accumulation strategy. I should have made that clear from the start.
Malcolm
Bill Bernstein
Posts: 604
Joined: Sat Jun 23, 2007 12:47 am

Re: Value Averaging

Post by Bill Bernstein »

As to the "inferior," we've been down this lane before; yes, the expected return of LS is less than DCA/VA, but only because it's riskier, as demonstrated by what happened to a lump sum investment in the US total stock market or S&P 500 starting at a market high, say, early 2000 or late 2007.

There's a much bigger point here, of course, which is that it's almost impossible for periodic young savers to "lose" with aggressive investing, whereas it's very possible, even over very long periods, for older investors with no human capital left, to "lose" with such a course.

Bill
umfundi
Posts: 3361
Joined: Tue Jun 07, 2011 5:26 pm

Re: Value Averaging

Post by umfundi »

wbern wrote:As to the "inferior," we've been down this lane before; yes, the expected return of LS is less than DCA/VA, but only because it's riskier, as demonstrated by what happened to a lump sum investment in the US total stock market or S&P 500 starting at a market high, say, early 2000 or late 2007.

There's a much bigger point here, of course, which is that it's almost impossible for periodic young savers to "lose" with aggressive investing, whereas it's very possible, even over very long periods, for older investors with no human capital left, to "lose" with such a course.

Bill
If you are a younger investor, invest what you have, when you can.

For the rest of what Dr. Bernstein says, for an older investor like me (who nevertheless denies his paucity of human capital), I will have to think about it. Bill is much wiser than I can hope to be.

Keith
Déjà Vu is not a prediction
Bill Bernstein
Posts: 604
Joined: Sat Jun 23, 2007 12:47 am

Oops

Post by Bill Bernstein »

Meant to say, of course, that the ER and risk of LS were both greater than with VA/DCA.

Bill
User avatar
LadyGeek
Site Admin
Posts: 71864
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Oops

Post by LadyGeek »

wbern wrote:Meant to say, of course, that the ER and risk of LS were both greater than with VA/DCA.

Bill
Some acronym decoding: "Meant to say, of course, that the Expected Return and risk of Lump Sum were both greater than with Value Averaging / Dollar Cost Averaging."

(ER is frequently used for Expense Ratio)

To the new investors: wbern is William Bernstein.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
bertilak
Posts: 8374
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Oops

Post by bertilak »

LadyGeek wrote:
wbern wrote:Meant to say, of course, that the ER and risk of LS were both greater than with VA/DCA.

Bill
Some acronym decoding: "Meant to say, of course, that the Expected Return and risk of Lump Sum were both greater than with Value Averaging / Dollar Cost Averaging."

(ER is frequently used for Expense Ratio)

To the new investors: wbern is William Bernstein.
With VA the risk can be driven either up or down depending on your current, actual, return. Risk is driven up by increasing equity allocation if the actual return is below your target return and is driven down if it is above target.

The raising and lowering of equity allocation (i.e. risk) is how VA attempts to force a target return. Of course there is no guarantee that you won't run out of money before the formula stops calling for more. In this sense VA is like doubling down in a casino.
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker (aka S.O.B.), the Cowboy Poet
umfundi
Posts: 3361
Joined: Tue Jun 07, 2011 5:26 pm

Re: Value Averaging

Post by umfundi »

mlewis wrote: Keith,
Again, just to clarify, my initial questions were not really with comparison to lump sum investing. I was asking them with respect to using VA as a long term accumulation strategy. I should have made that clear from the start.
Malcolm
Malcolm,

Systematic investing is indeed the way to go. Beyond that, I believe you should, as much as possible, put your plan on autopilot.

Value Averaging basically says that if the market goes up beyond your target, invest less aggressively, if it is below your target, invest more aggressively. Two obvious issues are: If it goes up and up, you will be relatively under-invested. If it goes down a lot, you will run out of cash to invest, so the theory is out the window.

The more subtle problem, I believe, is that you are continually having to make decisions about investing. There is no place I know that VA is a choice you can make, and put it on auto-pilot. Every pay period, you have to do some math and make a choice. I think this is too much opportunity for you to tamper with your strategy.

I think a preferable plan is to decide on a fixed Asset Allocation of, for example, the Three-Fund Portfolio, and then to make the systematic investments and to rebalance to maintain the AA.

Best wishes,

Keith
Déjà Vu is not a prediction
User avatar
telemark
Posts: 2866
Joined: Sat Aug 11, 2012 6:35 am

Re: Value Averaging

Post by telemark »

Controlling risk can mean different things. The common approach is to settle on some level of risk that you are willing to tolerate, and then hold that constant while trying to maximize returns. Value averaging takes the opposite approach: you determine the return needed to reach a specific goal, and then take on just enough risk to achieve that return.

So yes, I see it as controlling risk, but not in the way people normally think about it. If you have more than enough money to meet your needs and want to minimize the risk you take, value averaging is a way to fine-tune that.
User avatar
Topic Author
mlewis
Posts: 209
Joined: Mon Nov 05, 2012 12:39 am
Contact:

Re: Value Averaging

Post by mlewis »

Keith,
Good points. I can agree with that. I think DCA + rebalance is easier to implement and easier to stick with, and like you say, can leave less opportunity to tinker and make poor decisions.

Malcolm
User avatar
czeckers
Posts: 1049
Joined: Thu May 17, 2007 3:49 pm
Location: USA

Re: Value Averaging

Post by czeckers »

A VA strategy is executed at the portfolio level. By combining it with rebalancing, it will extend the strategy to the asset class level.

The crux of VA is plotting the expected value path. Too aggressive and you won't be able to add enough money to keep your portfolio on track. Too conservative and you end up with a significant percentage of your portfolio on cash -- potentially missing out on significant market returns. The more different asset classes you invest in, the more complicated this becomes. Even for a simple three fund portfolio, you'll have to figure the expected return of each asset class and then calculate the asset-weighted expected return of the whole portfolio. Also, if you shift your AA as you age, you'll have to recalculate every time your AA shifts.

The execution details are just that, details. To address the cash flow problem when significantly below the value path, you can always shift to a higher equity percentage until back on the value path -- over rebalancing if you will.

-K
The Espresso portfolio: | | 20% US TSM, 20% Small Value, 10% US REIT, 10% Dev Int'l, 10% EM, 10% Commodities, 20% Inter-term US Treas | | "A journey of a thousand miles begins with a single step."
SpartanlyPanly
Posts: 4
Joined: Tue Nov 27, 2012 9:52 am

Re: Value Averaging

Post by SpartanlyPanly »

mlewis wrote:A few questions on Value Averaging:

Do you think value averaging will increase returns or merely decrease risk? (although these are admittedly two sides of the same coin)

Would you value average your portfolio or the portfolio's individual components? I feel either method could pose its own theoretical problems.

Does long-term value averaging make sense for most people?
ad 1. I don't know. Depends a lot on how one defines and/or measures both.

Ad 2. I value average my portfolio. With the individual components, the (cash) management becomes very soon very complex.

Ad. 3. Probably not. As you can see from this thread, almost no one applied it but many have an opinion, some even without having read or understood the Edleson book.

A more important reason why it can't work for many people, is psychology.
Sitting back and doing nothing while waiting is hard enough for most people- esp. in a rising market with loads of 'opportunities' - but on top of that the strategy forces you to pour lots of cash in a seemingly bottomless hole, grabbing falling knifes while the whole world shouts "Don't!" Therefore, if you're not a contrarian, don't even think about it.

Furthermore, it requires a minimum of maths and even a formula, excel... beyond the treshold of "most people", which treshold can be put at zero.

Anyway, my experience:
I have made some changes to the Edleson strategy, having to do with treatment of dividends (to increase the next "target point" with the amount received in the current period or to keep the cash...?), with unexpected excess savings (am still struggling with this luxury problem) and the the choice of portfolio assets (not index trackers, more a long/short stock portfolio).

The psychological benefit of having pre-determined targets every time period and being forced to stick to them, is for me the big advantage of VA.
User avatar
czeckers
Posts: 1049
Joined: Thu May 17, 2007 3:49 pm
Location: USA

Re: Value Averaging

Post by czeckers »

To answer your excess return question: I had two different tax deferred accounts with identical AAs, both at VG. For 6 years (up until last July) I've been doing VA in one and DCA + rebalance with the other. The VA account's growth was 0.5% better per year, averaged across that period.

Contributions were still relatively large 0.5% of portfolio per month. It was a lot more work maintaining the VA account but I enjoy that sort of thing. Every month deposits go into MMF and dividends too. Quarterly I'd first check portfolio value against value path to determine whether to deploy cash, do nothing, or sell mutual funds. If action was required, I'd use purchases/sales to reset AA. The DCA goes in on autopilot and i rebalance on Halloween (to remind myself that the markets can be scary and thats ok) I think it becomes harder to keep on the value path as your contributions become relatively smaller and smaller. If your quarterly contribution is 3-6% of portfolio, then you can catch up after a 3-4% market drop. After a while though, you may have a great buying opportunity as in 2009 but not enough cash on hand to take advantage.

My employer had changed plans in July and I now have 3 diff accounts and pay $8 for any transaction beyond initial deposit so have gone to full DCA+rebalance because transaction fees can easily eat up the 0.5% advantage.

Hope this real world VA experience helps.

-K
The Espresso portfolio: | | 20% US TSM, 20% Small Value, 10% US REIT, 10% Dev Int'l, 10% EM, 10% Commodities, 20% Inter-term US Treas | | "A journey of a thousand miles begins with a single step."
Post Reply