Asymmetric Rebalancing: What and Why?

[Browser]

I've heard about a rebalancing strategy in which, I believe, the investor only rebalances when stocks have exceeded their allocation target, but not otherwise. It might be used with a rebalancing band, such as 10%. For example, if I have a portfolio with 50% bonds, 50% stocks, I would sell stocks to buy bonds only if my stock allocation had grown to 55%. I don't believe you ever sell bonds to buy stocks with this method, but I'm not sure. It's a "sell high" but not a "buy low" approach. I think Larry Swedroe mentioned in another thread somewhere that he uses this method with his personal portfolio. I'm wondering what the pros and cons are for using this approach, and what type of an investor it might be suitable for.

[EDN]

I've heard about a rebalancing strategy in which, I believe, the investor only rebalances when stocks have exceeded their allocation target, but not otherwise. It might be used with a rebalancing band, such as 10%. For example, if I have a portfolio with 50% bonds, 50% stocks, I would sell stocks to buy bonds only if my stock allocation had grown to 55%. I don't believe you ever sell bonds to buy stocks with this method, but I'm not sure. It's a "sell high" but not a "buy low" approach. I think Larry Swedroe mentioned in another thread somewhere that he uses this method with his personal portfolio. I'm wondering what the pros and cons are for using this approach, and what type of an investor it might be suitable for.

The pros? You'll feel good for a few months when stocks continue to fall and you didn't buy more. But when they recover, you'll have far fewer equities and recover much more slowly. Riding down a higher equity allocation but riding up a smaller one is not a good idea. You are better off just holding a slightly safer allocation and rebalancing both ways.

I spend my days trying to keep people from doing this out of fear. I can't image doing it as a matter of policy!

Eric

[Tonen]

For an accumulator this tends to happen in practice as you add new moneys. When stocks fall, new moneys tend to defray the need to rebalance out of bonds.

[umfundi]

Browser,

I have done my own studies that convince me timing when or how to rebalance is of no use. In fact, I think continuous rebalancing is the best strategy.

I know that some believe in "momentum", so you should not rebalance too soon. Those same people also believe in "reversion to the mean" which is that prices will revert to some historical trend.

I believe I am too stupid to take advantage of these phenomena.

Keith

[archbish99]

But shouldn't the converse be true as well? If stocks are high, new money will defray the need to rebalance into bonds?

I've seen a similar theory espoused on the idea that you never sell stocks at a loss -- you keep, say, 5 years of expenses in bonds as a floor, and hope that no bear market lasts longer than five years. In good years, you replenish the reserve back to five years; in bad years, you just take your withdrawals from that reserve. Smacks of mental accounting to me, but I can see where someone might feel happier not to be "throwing good money after bad."

[vesalius]

I get the impression Larry sees it as a viable alternative for those that have already won the game and have little to no reason to take additional risk. With stocks by far having risk of greatest loss and with Larry having all equities, only 30% or so of his money, in risky and sometimes volatile small, value and emerging stocks, this strategy keeps him from ever throwing money down a rabbit hole in extreme markets that can drop 50-80%. He will miss out on the potential upside if he had rebalanced into the market lows, but given those corrections can take many years and he has no need, why take the risk of rebalancing into a longer term, protracted bear market. It is all about marginal utility of wealth and the need to take risk. Similar to how he utilizes the fat-tail portfolio, he decreases both his chance of extreme losses and gains as a result, but, since he has enough money already, he really has no need to chance either of those.

[Call_Me_Op]

I don't think it's a terrible idea, and employed it myself in 2008 in my 401k. It is another method for risk control - and although it does likely reduce long-term return, it places a floor on losses. It is better suited to folks that are later in their accumulation years.

In 2008, I did continue to contribute to stocks weekly - but didn't sell bonds to buy stocks. Things have turned-out quite nicely.

Keep in mind that theoretically, a falling stock market can wipe-out almost all of your money even if you have (say) a 50-50 stock-bond allocation.

[jeffyscott]

I get the impression Larry sees it as a viable alternative for those that have already won the game and have little to no reason to take additional risk.

Since we now have enough, I have limits to how far I will go in adding to stocks in a decline for this reason.

We plan to retire early, this will require some extra funds for the years prior to beginning SS. Enough to cover that period will be kept in very low risk bonds/cash, no matter what. I'm not going to endanger my early retirement by chasing potential higher returns in stocks.

Later in retirement, our retirement security plan will have other constraints. These will limit the amount we would put in stocks in a decline. I did a (hopefully) worst case analysis and determined we need $X to ensure an acceptable minimum income, in any rebalancing the allocation to bonds/cash will not be allowed to fall below $X.

[EDN]

The whole notion of an asset allocation is to target a level of return that works for you commensurate with a given level of volatility. When you decide not to restore your equity allocation to its original percentage after a precipitous decline, you are choosing to change that profile for the worse--lower future returns while still participating (at a lower allocation) in future declines.

"Risk" is not the notion that equities may go down further before they recover. Risk is owning a portfolio whose growth will not keep pace with your spending needs or being forced to sell equity assets at temporarily depressed prices because you don't have enough in high-quality fixed income to sell instead. The answer to reduce both risks is to hold a moderately balanced portfolio (50-60% in stocks) and rebalance to and from. Much of the incremental return of that approach actually comes from buying more equities (with bond proceeds) at lower prices after a decline when future expected returns are above average. And the bigger the drop, the higher the expected return.

We can call "not rebalancing when we should" something more scientific of intentional like asymmetric rebalancing--but it's not. It's just what we've always referred to as market timing/bad behavior, etc.

If you have to do it (unable to overcome your own behavioral demons), so be it. But for goodness sake, don't plan on it ahead of time!

Eric

[Call_Me_Op]

Eric,

I am afraid I must disagree somewhat. There is more than one road to Dublin. Investors should be comfortable with whatever approach they are taking. As long as they understand that certain approaches have higher or lower expected returns and higher or lower risks, there is no reason they should not design an approach that suits their specific needs (including the need to sleep well at night). I do not agree that this should be called market timing or bad behavior any more than neglecting a certain asset class (like precious metals or REITs) should be considered bad behavior.

[EDN]

Eric,

I am afraid I must disagree somewhat. There is more than one road to Dublin. Investors should be comfortable with whatever approach they are taking. As long as they understand that certain approaches have higher or lower expected returns and higher or lower risks, there is no reason they should not design an approach that suits their specific needs (including the need to sleep well at night). I do not agree that this should be called market timing or bad behavior any more than neglecting a certain asset class (like precious metals or REITs) should be considered bad behavior.

I'm not sure "more than one road" applies here. That just means there isn't just one correct asset allocation (stock/bond or asset class breakdown). But there is an implicit assumption that whatever mix you choose, it will be rebalanced in good times and bad. All you are doing by not rebalancing when stocks are down is targeting a lower expected returning portfolio when expected returns are highest, lowering the total long-term expected return of your portfolio without a commensurate reduction in volatility (the losses already happened at a higher equity level). You'd be better off holding less stocks in the first place but actually rebalancing when they are down. Higher risk (riding down a higher-equity allocation), lower return (having less higher return assets when expected returns are highest) portfolios aren't a "road", they are a ditch you want to avoid on your travels.

As the saying goes, 'if you took the risk (associated with a certain stock allocation), you might as well get the return (by ensuring you hold that allocation when the market heads higher)'.

Eric

[vesalius]

Eric,

I am afraid I must disagree somewhat. There is more than one road to Dublin. Investors should be comfortable with whatever approach they are taking. As long as they understand that certain approaches have higher or lower expected returns and higher or lower risks, there is no reason they should not design an approach that suits their specific needs (including the need to sleep well at night). I do not agree that this should be called market timing or bad behavior any more than neglecting a certain asset class (like precious metals or REITs) should be considered bad behavior.

I disagree with Eric also. Continually rebalancing into a multiyear downturn in the market that would have them unnecessarily selling off bonds and eroding their nest egg. A young person can be reasonable sure he will recoup those equity and bond losses in the intermediate or long term. The retiree or near retiree not only may not have enough time to recoup but also has no reason to take that risk when they already have enough. The prospect of earning more to them in the future, potentially after they have died, means little to their present needs, which have already been met with current money already in hand.

[linuxizer]

I really don't buy the argument that not rebalancing during serious stock market crashes is "capitulation" and is as bad as selling everything, and doubly so if you've planned it. Are you giving up on some return this way? Likely so. But if it helps you sleep at night, this is about the most minor of all possible investing sins.

[jeffyscott]

"Risk" is not the notion that equities may go down further before they recover.

Risk is they don't recover in my lifetime. If recovery is certain, then there is no risk.

Risk is...being forced to sell equity assets at temporarily depressed prices because you don't have enough in high-quality fixed income to sell instead.

This is a reason for having constraints, such as keeping at least $X in high quality fixed income to cover needs for the next, say, 5-10 years. So I'd not rebalance, if it would reduce this to less than $X.

Let's say I have designated $X is $500,000 for me and I currently have $600,000 in high quality fixed income and $600,000 in stocks. Then stocks decline by 70%, to $180,000. Should I now return to 50/50 by reducing my high quality fixed income to $390,000 or should I keep at least $500,000 safe?

[umfundi]

Since we seem to be having a vote, I'm essentially with Eric.

Let's say I have designated $X is $500,000 for me and I currently have $600,000 in high quality fixed income and $600,000 in stocks. Then stocks decline by 70%, to $180,000. Should I now return to 50/50 by reducing my high quality fixed income to $390,000 or should I keep at least $500,000 safe?

What does your (written) investment plan say?

If it says you will have $500,000 in near-cash and the rest in some AA, then you should rebalance the remainder.

Eric's point is correct: If you have set an allocation and rebalance only some of the time, you are missing opportunities. In all probability the AA you have chosen is not really appropriate for you.

Keith

[Call_Me_Op]

Where is it written that rebalancing must be performed in a specific manner? I'd like to reserve the flexibility to be a bit more creative. If it is highest expected return you are shooting for, rebalancing isn't a factor because you should be 100% in equities. If the goal is to have enough to retire comfortably and to enjoy the ride along the way, I see nothing wrong with lopping the top off of your aggressive assets when they are doing very well, and letting them fend for themselves after they have taken a hit. Yes, you may be sacrificing some "expected return", but you are taking the prospect (however improbable) of total devastation off the table.

[Bacchus01]

In another thread, I think there was someone who calculated that, starting in early 2007 with a 75/25 portfolio, rebalancing on the way down and on the way up with a 5% band, that the end result was a few percentage points different than just holding that 2007 portfolio and riding it through.

[pkcrafter]

Eric,

I am afraid I must disagree somewhat. There is more than one road to Dublin. Investors should be comfortable with whatever approach they are taking. As long as they understand that certain approaches have higher or lower expected returns and higher or lower risks, there is no reason they should not design an approach that suits their specific needs (including the need to sleep well at night). I do not agree that this should be called market timing or bad behavior any more than neglecting a certain asset class (like precious metals or REITs) should be considered bad behavior.

I'm not sure "more than one road" applies here. That just means there isn't just one correct asset allocation (stock/bond or asset class breakdown). But there is an implicit assumption that whatever mix you choose, it will be rebalanced in good times and bad. All you are doing by not rebalancing when stocks are down is targeting a lower expected returning portfolio when expected returns are highest, lowering the total long-term expected return of your portfolio without a commensurate reduction in volatility (the losses already happened at a higher equity level). You'd be better off holding less stocks in the first place but actually rebalancing when they are down. Higher risk (riding down a higher-equity allocation), lower return (having less higher return assets when expected returns are highest) portfolios aren't a "road", they are a ditch you want to avoid on your travels.

As the saying goes, 'if you took the risk (associated with a certain stock allocation), you might as well get the return (by ensuring you hold that allocation when the market heads higher)'.

Eric

Eric is correct,+1.


Paul

[EDN]

OK,

Let me try to address as many of the comments as I can:

1. It appears like different posters are bringing different circumstances to the table. If you have "won" or "made it", that means you have more money than you will ever be able to spend and don't have any 'beyond the grave' objectives. In that circumstance, the safest thing to do is simply to buy a SPIA and be done with it. So I don't think this situation even applies to the discussion.

2. I am assuming you hold some combination of stocks and bonds because you need future portfolio growth to compliment a rising income stream (you haven't "won" but have a reasonable amount of assets for your objectives) and further may wish to leave an inheritance/bequest. Now, I agree that failing to rebalance during the most severe declines is not as bad as bailing out of stocks completely, but that doesn't mean it is OK or preferrable as a policy, and that is my point (the goal is not lower returns for a given level of risk, which is what this would accomplish).

3. On the topic of multi-year bear markets for equities (we've seen 3 in 85 years: '29-'32, '73-'74, and '00-'02) and continuing to rebalance, I still don't believe most retirees not in situation #1 should just sit on their hands. Why? Well, in the case mentioned of a -70% decline (that's only happened once in 85 years in the US), the decline has already happened. The only question is at that point, do you remain at the lower-expected-returning allocation (less in stocks) or do you return to your original plan at a time when the expected return of that allocation is higher and the higher returns are needed to bring the plan back in line with original target? Let's consider the options:

a) It is most likely that at that level of decline, you will outlast your money if you keep the vast majority of your future assets in lower returning bond investments and do not restore your portfolio to its original target.
b) It is possible that your equities will either stay at these low levels, or the higher-expected returns do not materialize (in which case rebalancing into them doesn't hurt, it just doesn't help as much). This has not happened, but anything's possible.
c) it is probable that the higher expected returns on equities will materialize, and restoring your portfolio to its original allocation will help to offset its temporary set-back and you will be able to ultimately achieve your objectives. This is what always has happened, so you're making a big bet against reasonable assumptions and a lot of history if you ignore it.
d) there is always a chance that all stocks in the world go to 0 and you will have lost more money (permanently) by continuing to rebalance -- however this is such an extreme assumption I'm not sure you'll be much better off without rebalancing (will $ even have value?)

=====================================

So what I am sensing is that we like the potential return of stocks with some part of our portfolio, just so long as they don't fall in value too much (and expected returns go up by too much) to make us uncomfortable. Unfortunately, that these declines have happened and probably will happen again is what allows stocks to achieve the returns they have and are expected to do.

But at the point of these declines, we begin to worry (irrationally?) that they will never recover or never come back in our lifetime, despite the evidence to the contrary (the real value on the market portfolio was fully restored in less that 5 years from the trough during '29-'32, '37, '73-'74, '00-'02, and '08, especially if we rebalanced and sold bonds to buy more stocks after they had declined) and despite the probability that the higher-bond allocation left alone will not have a sufficient value or expected return to achieve our goals. Put another way, we are trading the relative certainty that a lower risk/lower return will eventually cause us to run out of money with a very high probability for the remote chance that our plan was flawed from the start and actually rebalancing $ back to equities when they are down (to restore our original allocation) at higher-expected (and needed) returns just causes us to lose such an irreparable amount of money in the short-run that our plan blows up.

I'll leave it up to everyone to decide what is right for them. But my opinion is you invest the time and research into your plan for these very situations -- so you have the confidence to stay on course because that course is what offers you the highest probability of success (and all the bumps and bruises that come along with it). If deviating from that plan will work better for you, it wasn't the right plan to begin with.

Most investors need to capture the entire return of their chosen allocation (and chosen level of risk) to achieve their goals. For the vast majority, earning a lower return (holding less in stocks when expected returns are highest) for a given level of risk (holding the most in stocks when stocks are at their highest prices and lowest expected returns) simply won't allow them to be successful. It may feel comfortable at the time to make this decision, but avoiding some short-term pain will likely lead to much more down the road.

Thats about all I can say on the matter.

Eric

[grayfox]

Rebalancing to a constant mix is not the only dynamic strategy. See the 1994 paper Dynamic Strategies for Asset Allocation by Andre F. Perold and William F. Sharpe.

Another method described in the paper is simple Buy and Hold, i.e. no-rebalancing. In Figure 4, Buy-and-Hold is shown to be superior to Constant Mix in trending markets.

Sharpe won a Nobel prize, so I would give some weight to his ideas. In my opinion, unyielding dogmatism about one strategy or another is not helpful.

[Browser]

I believe Jack Bogle has said he doesn't believe in rebalancing at all. The asymmetric strategy seems to be half-way based on that premise, since it doesn't rebalance except when equities have exceeded the policy target. This has to have the effect of allowing portfolio risk to drift lower when stocks have declined in value, but not drift much higher when stocks are increasing in value. I would think this offers more downside (left tail) protection, at the cost of lowering the upside (right tail) potential. Seems consistent with Swedroe's "minimize fat tails" portfolio strategy. It would be more suitable for investors who are more sensitive to left tail events, such as retirees in distribution stage.

[jeffyscott]

What does your (written) investment plan say?

Why does that matter, if I wrote down something that I should not do, does that make it okay? What if I write down that when stocks exceed their 200 day moving average, I will switch to 100% stocks, in order to catch the momentum?

If it says you will have $500,000 in near-cash and the rest in some AA, then you should rebalance the remainder.

Eric's point is correct: If you have set an allocation and rebalance only some of the time, you are missing opportunities.

If you say you are missing opportunities, then you are saying there is no risk. You can say you may be missing opportunities (but you also may be missing a disaster).

Also it appears that rebalancing only some of the time is okay, as long as I have written it down. I have $600,000 in fixed income and $600,000 in stocks, my allocation is 50/50. Stocks drop by 1/3 and I rebalance to $500,000 stocks and $500,000 fixed income. But it's like 1929 and stocks continue to fall, I don't rebalance any further, my $500,000 in stocks drops to $100,000, but I don't change anything, so I end up with $100,000 in stocks and $500,000 (83%) fixed income.

Now it may be from that low point stocks soon skyrocket and I will have missed an opportunity, but it also may be that stocks have reached a permanently low basin and I have avoided a worse disaster.

[jeffyscott]

But at the point of these declines, we begin to worry (irrationally?) that they will never recover or never come back in our lifetime, despite the evidence to the contrary (the real value on the market portfolio was fully restored in less that 5 years from the trough during '29-'32, '37, '73-'74, '00-'02, and '08,...

So this can't happen here?

[Browser]

I've always questioned the idea that a fixed asset allocation has fixed risk, and you are maintaining that risk level by rebalancing. I don't believe it. For example, when stocks have declined in price it seems to me that's because the market has decided that the current risk premium for owning stocks is too low; i.e., stock risk has increased relative to expected returns. So investors drive the price down to a level that restores the equilibrium risk premium. If you do nothing, the risk level of your portfolio remains the same as before, at least in terms of the market's judgment of risk. If you choose to buy more stocks ("rebalance") then you are increasing the risk level of your portfolio relative to the market's judgment of risk. So, rebalancing back into stocks after stocks have lost value is really a risk-reward story. You are assuming greater risk in the expectation of greater reward. But like all risk-reward stories, it wouldn't be risk if it always paid off. Even if it has often paid off in the past, there is the real, true, and actual possibility that it won't this time and you will get kicked in your tender spot.

[tfb]

I've heard about a rebalancing strategy in which, I believe, the investor only rebalances when stocks have exceeded their allocation target, but not otherwise. It might be used with a rebalancing band, such as 10%. For example, if I have a portfolio with 50% bonds, 50% stocks, I would sell stocks to buy bonds only if my stock allocation had grown to 55%. I don't believe you ever sell bonds to buy stocks with this method, but I'm not sure. It's a "sell high" but not a "buy low" approach. I think Larry Swedroe mentioned in another thread somewhere that he uses this method with his personal portfolio. I'm wondering what the pros and cons are for using this approach, and what type of an investor it might be suitable for.

I would go the opposite. Always buy low but slow to sell and let it ride.

[bertilak]

I've always questioned the idea that a fixed asset allocation has fixed risk, and you are maintaining that risk level by rebalancing. I don't believe it. For example, when stocks have declined in price it seems to me that's because the market has decided that the current risk premium for owning stocks is too low; i.e., stock risk has increased relative to expected returns. So investors drive the price down to a level that restores the equilibrium risk premium. If you do nothing, the risk level of your portfolio remains the same as before, at least in terms of the market's judgment of risk.

I have the same suspicion and have brought it up elsewhere but I think you have articulated it more clearly!

I might add that elsewhere it is taken as a given that "the market's judgement", although fickle, is the best estimate we have for value. Is it not also the best estimate we have for level of risk?

Don't try to fight the rise and fall of the tide. Your constant risk is in relation to sea level, not an absolute number.

[grayfox]

I've always questioned the idea that a fixed asset allocation has fixed risk, and you are maintaining that risk level by rebalancing. I don't believe it. For example, when stocks have declined in price it seems to me that's because the market has decided that the current risk premium for owning stocks is too low; i.e., stock risk has increased relative to expected returns. So investors drive the price down to a level that restores the equilibrium risk premium. If you do nothing, the risk level of your portfolio remains the same as before, at least in terms of the market's judgment of risk. If you choose to buy more stocks ("rebalance") then you are increasing the risk level of your portfolio relative to the market's judgment of risk. So, rebalancing back into stocks after stocks have lost value is really a risk-reward story. You are assuming greater risk in the expectation of greater reward. But like all risk-reward stories, it wouldn't be risk if it always paid off. Even if it has often paid off in the past, there is the real, true, and actual possibility that it won't this time and you will get kicked in your tender spot.

Good insight.

It is well known that volatility varies over time. Why wouldn't the perceived risk also vary over time?

Say you invest in something, say, oil trade and the price is P and the expected return is 10%. Suddenly there is news that someone is threatening to sink oil tankers in some narrow strait. Hasn't the perceived risk increased? The market may lower the price to P/2 so that expected return is 20%, to compensate for the increased perception of risk.

[Browser]

I've heard about a rebalancing strategy in which, I believe, the investor only rebalances when stocks have exceeded their allocation target, but not otherwise. It might be used with a rebalancing band, such as 10%. For example, if I have a portfolio with 50% bonds, 50% stocks, I would sell stocks to buy bonds only if my stock allocation had grown to 55%. I don't believe you ever sell bonds to buy stocks with this method, but I'm not sure. It's a "sell high" but not a "buy low" approach. I think Larry Swedroe mentioned in another thread somewhere that he uses this method with his personal portfolio. I'm wondering what the pros and cons are for using this approach, and what type of an investor it might be suitable for.

I would go the opposite. Always buy low but slow to sell and let it ride.

I'm wondering if this version of asymmetric rebalancing would be more suitable for accumulators and the Swedroe version for decumulators, since basically your approach effectively increases average portfolio risk over time and the other decreases it, both relative to a fixed allocation. Or, you could just increase your equity allocation to increase the risk and rebalance the old-fashioned way.

[Browser]

It is well known that volatility varies over time. Why wouldn't the perceived risk also vary over time?

Yes, and it's also well known that stock volatility is higher when stocks stocks are going down vs. going up. This would be consistent with the notion that stocks (in terms of volatility) are getting riskier when they go down in price.

[umfundi]

There is a missing component here: What are you rebalancing with? Is this like the arguments about Dollar-Cost Averaging, where the whole focus is on the "risk" of stocks, but the other part is some mythical safe liquid cash reserve that has no risk at all?

I ran some simulations some time ago. The scenario was a 50/50 portfolio of stocks and bonds (TSM and TBM) with a monthly investment for about a dozen years to the start of 2012.

Case 1: Monthly investment equally in stocks and bonds, no rebalancing.
Case 2: Monthly investment of the same amount, portfolio rebalanced to 50/50 each month.

The result was enough to convince me that monthly rebalancing is worth a bonus of about 0.3% per year in returns vs. not rebalancing.

Given that, it would seem that not rebalancing some of the time can only cut the rebalancing bonus. I cannot see that it makes any difference if you rebalance only on the upside, or only on the downside.

And, I tend to agree with comments by others, that elaborate rebalancing strategies are market timing (or, even worse, technical analysis) in disguise.

Keith

[grayfox]

I ran some simulations some time ago. The scenario was a 50/50 portfolio of stocks and bonds (TSM and TBM) with a monthly investment for about a dozen years to the start of 2012.

There's the answer right there. 2000-2012 has been a non-trending period with big reversals-down, up, down, up--ending up about where it started. Re-balancing outperforms Buy-and-Hold in trendless markets.

If you picked a trending period, my guess is that instead of re-balancing bonus, there would be a re-balancing penalty.

[Browser]

There is a missing component here: What are you rebalancing with? Is this like the arguments about Dollar-Cost Averaging, where the whole focus is on the "risk" of stocks, but the other part is some mythical safe liquid cash reserve that has no risk at all?

I wondered about that too. If you have a stock/bond portfolio you are, of course, rebalancing between those two. Asymmetric rebalacing simply sets different "bands" above and below the equity allocation, I believe. With Swedroe's approach, you might have a 5% band above the target equity allocation no band below the target allocation (you never rebalance if equities are below target). Or you could do what tfb suggests: a 5% band below the target equity allocation and no band above (you never rebalance as long as the equity allocation stays above target). Or you might have 5% below, 10% above, for example, which lets the equity gains run somewhat but has an upper ceiling.

[letsgobobby]

I've heard about a rebalancing strategy in which, I believe, the investor only rebalances when stocks have exceeded their allocation target, but not otherwise. It might be used with a rebalancing band, such as 10%. For example, if I have a portfolio with 50% bonds, 50% stocks, I would sell stocks to buy bonds only if my stock allocation had grown to 55%. I don't believe you ever sell bonds to buy stocks with this method, but I'm not sure. It's a "sell high" but not a "buy low" approach. I think Larry Swedroe mentioned in another thread somewhere that he uses this method with his personal portfolio. I'm wondering what the pros and cons are for using this approach, and what type of an investor it might be suitable for.

I would go the opposite. Always buy low but slow to sell and let it ride.

This is what I do. My AA is 60/40 but I buy stocks with new money on RBDs even if I'm not quite at 55/45. I don't sell unless I'm at 65/35 or worse (although i did last year because my international/domestic was out of rebalancing band). Furthermore I incoporate PE10 in my IPS so that contraction below historic means is acted on more rapidly than expansion above. Otherwise 100% of tax-deferred money goes to bonds, and 100% of taxable money goes to I bonds and stocks, and this basically means my AA only changes when stocks make a big move up or down.

In essence I think this means my AA is really higher than 60/40, but I just don't call it that.

when stocks are rising, I rebalance to maintain risk.

when stocks are falling, I rebalance to maintain expected return.

They are both sides of the same coin.

[umfundi]

I ran some simulations some time ago. The scenario was a 50/50 portfolio of stocks and bonds (TSM and TBM) with a monthly investment for about a dozen years to the start of 2012.

There's the answer right there. 2000-2012 has been a non-trending period with big reversals-down, up, down, up--ending up about where it started. Re-balancing outperforms Buy-and-Hold in trendless markets.

If you picked a trending period, my guess is that instead of re-balancing bonus, there would be a re-balancing penalty.

I picked the period because that is the data I could find. I was also trying to figure out why I had done just fine, thank you, during this "lost decade".

It is well known that not rebalancing over long periods is advantageous because your Asset Allocation drifts to riskier investments with higher returns.

Rebalancing takes advantage of the noise between investment classes that are not perfectly correlated while maintaining your overall portfolio risk in terms of AA. It is not a technique to capitalize on trends.

As I said in another thread, I am too stupid to know whether Monday will be a momentum day or a reversion to the mean day. So, my bias is towards frequent or continuous rebalancing, which I get in the Vanguard Lifecycle Strategy funds.

Keith

[umfundi]

In essence I think this means my AA is really higher than 60/40, but I just don't call it that.

Now, that's a great insight. Really!

Let's say your IPS allocation is 60/40% stocks/bonds. But, you set your rebalancing bands so you will rebalance when stocks are 66% on the upside and 58% on the downside. How is that materially different than a 62% IPS allocation with equal +-4% bands?

(I just got a notice of increased fees from my mental accountant!)

Keith

[larryswedroe]

I was asked to comment and here are few thoughts, though I don't have time to read the whole thread

First, in absence of frictions (taxes and trading costs) one should rebalance daily. Simple logic tells you that should be the case. There is no reason to allow market to determine the amount of risk you are taking. Swensen of Yale follows this approach, using cash flows to rebalance

Second, in real world we have frictions so some straying should be allowed. There is actually science/math behind the optimal strategy and it is one of the most difficult problems in finance especially with taxable accounts and different lots with different basis. There are actually programs out there now commercially that many advisors use and IMO they are all bad as they ignore the frictions, at least ones I have seen. We actually have been working on this problem for several years and in combination with another firm we believe we have solved the math problem and will begin using it in very near future, instead of the reasonable 5/25 rule

Third, I choose asymmetric rebalancing for simple reason that I have no need to take risk and want to sleep well so I can enjoy life and not worry about especially since my ownership interest in my firm has large equity exposure. I would not suggest that it is the right strategy for others. But it helps me since I have zero need for risk but still choose to take some as there are periods when stocks do well and bonds poorly. But the level is such that if the stocks went to zero it would not change my lifestyle at all.

I do rebalance on the upside though for stocks.

I hope that is helpful
Larry

[Rodc]

First seems to me is to define one's goals. Say you really want $200K four years from now. Or you are trying to a generate a specific fixed income from a portfolio, as opposed to say being 40 and all you are trying to do is grow your portfolio and you don't yet have very specific goals.

The generic set an allocation by percent in stocks and bonds and rebalance to maintain a (roughly) set level of risk might not line up very well with the actual risk of not meeting goals. A fixed amount in safe assets may better fit the goals. In this case it makes sense to rebalance only from stocks to bonds (TIPS, etc).

Does this mean you give up tons of returns as many argue? First, let us remember that a ton of returns might not be the goal! This seems entirely forgot in posts like the very first reply. The goal might be to make sure you do not lose money, because you already have enough.

Ok, a simple example. You start with a stock index and a bond index both at 100. You have $100K in stock and bonds funds tied to these indexes.

At the end of year 1 the stock index is 75, bonds no change. You sell bonds to buy stocks.

At the end of year 2 the stock index is 50, bonds no change. You sell bonds to buy stocks.

At the end of year 3 stocks are back to 75, bonds no change. You sell stocks to buy bonds.

At the end of year 4 stocks are back to 100, you sell stocks to buy bonds.

You have about $212K. If you do not rebalance you have $200K.

Rebalancing helped, maybe enough to matter to you, or not. It is not enough most likely to change your life.

But there was increased risk for someone who needed that money in one of those years (say this was a college fund, or a retiree who needed to generate cash income): there was no guarantee the market was going to snap back. The after the last crash the market did snap back, but it could easily have taken far longer to come back. Indeed it may not yet be over. Have to wait and see how things go.

If you have enough, and protecting your assets is more important than growing your assets, rebalancing one way is a very reasonable thing to do, and the standard one size fits all rebalancing advice just might not fit.

[Browser]

I think I pretty much agree with Bogle about rebalancing - it ain't worth worrying about for most of us unless things get really out of line. When the market dropped by 50% in 2008 it might have been a good time to rebalance into stocks. When the market went nuts in 1995-1999, it might have been a good time to rebalance out of stocks and into bonds.

We’ve just done a study for the NYTimes on rebalancing, so the subject is fresh in my mind. Fact: a 48%S&P 500, 16% small cap, 16% international, and 20% bond index, over the past 20 years, earned a 9.49% annual return without rebalancing and a 9.71% return if rebalanced annually. That’s worth describing as “noise,” and suggests that formulaic rebalancing with precision is not necessary.

We also did an earlier study of all 25-year periods beginning in 1826 (!), using a 50/50 US stock/bond portfolio, and found that annual rebalancing won in 52% of the 179 periods. Also, it seems to me, noise.

http://seekingalpha.com/article/41119-vanguard-s-jack-bogle-on-rebalancing-don-t

[The Wizard]

Now it may be from that low point stocks soon skyrocket and I will have missed an opportunity, but it also may be that stocks have reached a permanently low basin and I have avoided a worse disaster.

So, like, when was the last time stocks reached a permanently low basin, to use that quaint terminology?
It hasn't happened and it's not GOING to happen. So ignore that fantasy...

[Browser]

Now it may be from that low point stocks soon skyrocket and I will have missed an opportunity, but it also may be that stocks have reached a permanently low basin and I have avoided a worse disaster.

So, like, when was the last time stocks reached a permanently low basin, to use that quaint terminology?
It hasn't happened and it's not GOING to happen. So ignore that fantasy...

You have obviously avoided the Japanese market since 1989, as well as the markets in many countries in which the stock markets completely collapsed and investors lost everything. But that can't happen here...

[The Wizard]

Now it may be from that low point stocks soon skyrocket and I will have missed an opportunity, but it also may be that stocks have reached a permanently low basin and I have avoided a worse disaster.

So, like, when was the last time stocks reached a permanently low basin, to use that quaint terminology?
It hasn't happened and it's not GOING to happen. So ignore that fantasy...

You have obviously avoided the Japanese market since 1989, as well as the markets in many countries in which the stock markets completely collapsed and investors lost everything. But that can't happen here...

I live and mostly invest in the US, correct.
So I'm dealing with that reality, not worrying about boogyman scenarios.
Buy low, sell high. Rinse & repeat....

[Browser]

I live and mostly invest in the US, correct.
So I'm dealing with that reality, not worrying about boogyman scenarios.
Buy low, sell high. Rinse & repeat....

Gee, that doesn't sound ethnocentric at all.

[jsl11]

Most of the discussion here about 2008 ignores the fact that this was not an ordinary stock market decline. Major financial companies were failing. Merrill Lynch essentially went broke. Bear Stearns and Lehman Bros. went belly up. The government had to bail out AIG and many (most?) of the banks. General Motors went bankrupt! The US was on the verge of financial collapse. It was not at all clear how it would all be resolved, nor how long it would take. A stock market recovery seemed to be a remote possibility any time soon. For a retiree who had "enough", selling bonds to buy stocks did not seem to be a smart thing to do. At that time, such a retiree was much more concerned with asset preservation rather than possible, maybe, asset growth at some unknown time in the future. With hindsight, it is easy to say now that it would have been better to rebalance. At that time, it was not so clear.

Jeff

[magician]

Rebalancing to a constant mix is not the only dynamic strategy. See the 1994 paper Dynamic Strategies for Asset Allocation by Andre F. Perold and William F. Sharpe.

Another method described in the paper is simple Buy and Hold, i.e. no-rebalancing. In Figure 4, Buy-and-Hold is shown to be superior to Constant Mix in trending markets.

Sharpe won a Nobel prize, so I would give some weight to his ideas. In my opinion, unyielding dogmatism about one strategy or another is not helpful.

I recall writing that somewhere, probably in the Do You Rebalance? thread.

[Leesbro63]

Most of the discussion here about 2008 ignores the fact that this was not an ordinary stock market decline. Major financial companies were failing. Merrill Lynch essentially went broke. Bear Stearns and Lehman Bros. went belly up. The government had to bail out AIG and many (most?) of the banks. General Motors went bankrupt! The US was on the verge of financial collapse. It was not at all clear how it would all be resolved, nor how long it would take. A stock market recovery seemed to be a remote possibility any time soon. For a retiree who had "enough", selling bonds to buy stocks did not seem to be a smart thing to do. At that time, such a retiree was much more concerned with asset preservation rather than possible, maybe, asset growth at some unknown time in the future. With hindsight, it is easy to say now that it would have been better to rebalance. At that time, it was not so clear.

Jeff

I disagree. It's always scary. In 1929 was THAT a normal decline? In 1974, when cars got 8MPG and oil looked to be running out in 16 years and prices more than doubled overnight....was THAT a normal decline?

[telemark]

Most of the discussion here about 2008 ignores the fact that this was not an ordinary stock market decline. Major financial companies were failing. Merrill Lynch essentially went broke. Bear Stearns and Lehman Bros. went belly up. The government had to bail out AIG and many (most?) of the banks. General Motors went bankrupt! The US was on the verge of financial collapse. It was not at all clear how it would all be resolved, nor how long it would take. A stock market recovery seemed to be a remote possibility any time soon. For a retiree who had "enough", selling bonds to buy stocks did not seem to be a smart thing to do. At that time, such a retiree was much more concerned with asset preservation rather than possible, maybe, asset growth at some unknown time in the future. With hindsight, it is easy to say now that it would have been better to rebalance. At that time, it was not so clear.

Jeff

Times like that are exactly the reason you want to have a written plan that you can stick to. It always feels like this time is different. Rebalancing according to a plan is one of the best ways to avoid behavioural errors.

[roymeo]

Where is it written that rebalancing must be performed in a specific manner? I'd like to reserve the flexibility to be a bit more creative. If it is highest expected return you are shooting for, rebalancing isn't a factor because you should be 100% in equities. If the goal is to have enough to retire comfortably and to enjoy the ride along the way, I see nothing wrong with lopping the top off of your aggressive assets when they are doing very well, and letting them fend for themselves after they have taken a hit. Yes, you may be sacrificing some "expected return", but you are taking the prospect (however improbable) of total devastation off the table.

Uh, I believe I've seen studies that show that the reality is that having an un- or low-correlation other investment category will increase returns.

[umfundi]

I rebalanced through 2008. More precisely, my (low cost passive) advisor did it for me. Not only that, I retired on 1 Nov 2008.

The efficient frontier moves around, but the graph I like to look at says minimum risk (volatility) is at about 20/80% stocks/bonds. Also, 40/60% is about the same risk as 0/100%, with substantially greater returns.

Keith

[jsl11]

Most of the discussion here about 2008 ignores the fact that this was not an ordinary stock market decline. Major financial companies were failing. Merrill Lynch essentially went broke. Bear Stearns and Lehman Bros. went belly up. The government had to bail out AIG and many (most?) of the banks. General Motors went bankrupt! The US was on the verge of financial collapse. It was not at all clear how it would all be resolved, nor how long it would take. A stock market recovery seemed to be a remote possibility any time soon. For a retiree who had "enough", selling bonds to buy stocks did not seem to be a smart thing to do. At that time, such a retiree was much more concerned with asset preservation rather than possible, maybe, asset growth at some unknown time in the future. With hindsight, it is easy to say now that it would have been better to rebalance. At that time, it was not so clear.

Jeff

I disagree. It's always scary. In 1929 was THAT a normal decline? In 1974, when cars got 8MPG and oil looked to be running out in 16 years and prices more than doubled overnight....was THAT a normal decline?

Well, I can remember 1974. I did not think it was scary. I was just unhappy that my holdings went down and did not rebound for quite some time. There certainly was no talk of the investment industry giants going out of business. We were not on the verge of a financial collapse. That bear market was well within the range of normalicy.
While I can't remember 1929, that was even worse than 2008. At that time, most people could not invest because:
1. Most people could not afford to invest.
2. There were no inexpensive ways to invest.
3. Most of those who invested were doing it on margin; it was the thing to do. Those investors lost everything.
4. If you wanted to invest, your money was lost or tied up when your bank failed.
5. There was 25% unemployment. Most people were worried about keeping/finding employment and putting food on the table.
There was no social security, no FDIC. People were selling pencils on street corners and standing in soup lines just to survive.
Rebalancing was so far beyond their concerns, that the mere mention of it would have been laughable.

Jeff

[grayfox]

Rebalancing to a constant mix is not the only dynamic strategy. See the 1994 paper Dynamic Strategies for Asset Allocation by Andre F. Perold and William F. Sharpe.

Another method described in the paper is simple Buy and Hold, i.e. no-rebalancing. In Figure 4, Buy-and-Hold is shown to be superior to Constant Mix in trending markets.

Sharpe won a Nobel prize, so I would give some weight to his ideas. In my opinion, unyielding dogmatism about one strategy or another is not helpful.

I recall writing that somewhere, probably in the Do You Rebalance? thread.

Yes, I was going to simply quote your post from that thread, but I didn't want to post your quote out of context and drag you into it.