Do you Rebalance ?

[tphp99]

Did not want to hijack the thread on rebalancing after 50% market drop.

So... what's the point of rebalancing again? Especially in the accumulation stage.

To rebalance is to time the market, and since I'm certain that my ability to time the market is horrible, I just don't rebalance. But who knows, maybe the next down turn or run-up will change that thinking.

What are your thoughts?

[BBL]

To rebalance is to time the market

No.

To rebalance is to return to your preselected risk profile in your AA as described in your IPS.

Think portfolio risk management not market timing.

I just don't rebalance.

The market is driving your risk - not your IPS. Not good.

[l2ridehd]

Re-balancing is the single most effective way to make your portfolio grow faster. I guess this also means you have not done the other most important thing, establish your best AA. So step one is to determine your AA. If not able to do that then just select the age old 100 minus age in bonds. Once that is done re-balance when ever your 5% out of balance or select a date specific and do it. Why does re-balancing grow your investments faster? Because it forces you to buy low and sell high. If you can't do it then invest in target date funds that do it automatically for you.

[dbr]

The purpose is to control the risk of your portfolio. In general risky assets should on average return more than less risky assets so over time the portfolio will become dominated by riskier assets. At some point the investor would want to get back to a more comfortable risk level. The market can move the other way and leave one with less than desired in risky assets, which can also be corrected, but the fundamental rationale is to prevent drifting to too high a risk profile.

In my opinion there is way too much angst over what is a very simple concept.

[gt4715b]

Like everybody is saying, it keeps the risk/return profile of your investments stable. This of course assumes that you have a target asset allocation. Another good reason for having a fixed, rule-based rebalancing is that is stops you from tinkering and/or overthinking your portfolio decisions in volatile markets. You follow the rules and get on with your life.

Of course, if you're in the early accumulation stage your new contributions should pretty much rebalance you automatically except in a volatile year like 2008/2009.

[tphp99]

As a risk management tool? Grows the whole portfolio faster?

None of it makes sense. How is my portfolio less risky if I rebalance when stocks go down? Wouldn't it put more money at stake?

Suppose the market drops 10% a day for 22 days straight and the bond market stays the same. My portfolio would be worse off if I rebalanced daily during this unlikely but possible event.

Just trying to learn, not being argumentative at all.

Thank you,

TP

[bottomfisher]

Periodic rebalancing according to your plan forces you into selling high and buying low assets without consciously attempting to market time.

[sscritic]

Why does re-balancing grow your investments faster? Because it forces you to buy low and sell high.

During a period of rising (stocks or bonds), it forces you to sell too soon. That may be high, but it is not the high. Rather than get the full benefit of the increase, you keep taking money off the table. I am not saying that rebalancing is bad, but this idea that rebalancing insures that you sell at a high is a myth. You sell at higher than something, but you are also selling at lower than something, and that higher than something might not be that high a higher and is certainly not the high.

[sscritic]

P.S. When both bonds and stocks were crashing in 2008-2009, if you sold bonds to buy stocks (which were crashing harder than bonds), you were selling bonds low, not high. They may have been higher than something, but they were very clearly lower than a lot of things.

[nydad]

True, but if anyone could actually sell at the high, they would just rebalance at the high... Since we don't know the high in advance, rebalancing helps capture some of the gain and maintain the risk profile.

[chrisj]

Here is an academic view of rebalancing with specific advice of how to implement a rebalancing strategy. I had to read it a couple of times for it to sink in.

http://www.tdainstitutional.com/pdf/Opportunistic_Rebalancing_JFP2007_Daryanani.pdf

[sscritic]

True, but if anyone could actually sell at the high, they would just rebalance at the high... Since we don't know the high in advance, rebalancing helps capture some of the gain and maintain the risk profile.

But you could adopt a strategy of waiting until a high is achieved and sell once it starts dropping. That could be higher or lower than selling on the way up, but rebalancing forces you to sell on the way up, unless by accident the day you rebalance is the high.

I agree completely on the risk profile part. Rebalance to keep your risk where you want it to be. My point is that a person should actually think about what "forces you to sell high and buy low" really means rather than just repeat it because someone else said it and it sounded cute.

[l2ridehd]

During a period of rising (stocks or bonds), it forces you to sell too soon. That may be high, but it is not the high.

sscritic, please the next time you know the exact high and the exact low, tell me. Until then I will re-balance when I am out of balance by 5%. And that will force me to sell high and buy low. Agree it might not be "the high" or "the low" but it will be far better then not doing it.

[nisiprius]

This is a frequently-argued topic. I just want to throw one thing in here. It really doesn't help things to label practices as market timing. If you look at what people who call themselves "market timers" do, what market timing systems do, what market timing newsletters do, it is very clear that this kind of market timing involves large sudden changes in asset allocation--often the complete sell-off of an asset class--trading in and out, as often as several times a year, based on technical market indicator "signals." Relationships between moving averages, perhaps.

In effect in this forum there are four kinds of rebalancers.

1) Those who invest in all-in-one funds like target retirement funds, that are automatically rebalanced daily. They don't fuss about it. They figure if rebalancing is important, it's being taken care of.

2) Those who rebalance "for risk control" rarely, often not even annually, and feel that you only need to rebalance when (say) your allocations have drifted, say, from something resembling LifeStrategy Conservative (40% stocks) to something resembling LifeStrategy Moderate (60% stocks). They have a casual attitude. It's a chore that needs to be done occasionally but it isn't of huge importance.

3) Those who rebalance rigorously according to a predefined system, perhaps including "rebalancing bands;" object to being called "buy-and-hold" investors and insist that they are practicing "buy, hold, and rebalance;" and believe that rebalancing is a source of worthwhile extra return, sometimes called a "rebalancing bonus." The rebalancing bonus exists to the extent that asset classes exhibit an active tendency to "revert to the mean." The people who do this tend to be "slice-and-dicers," and among other things tend to invest in a larger number of relatively volatile asset classes, whose percentage in the portfolio will fluctuate off target fairly quickly if not rebalanced.

Notice that if the reversion-to-mean rebalancing bonus exists, the automatic daily rebalancing of an all-in-one fund will not capture it.

If this is "market timing," it is very different from what people who call themselves "market timers" do. It does not involve large-commitment shifts in an out of asset classes, just small adjustments, and it isn't based on technical analysis.

4) Those who occasionally indulge in "opportunistic rebalancing." They still just rebalance back to their target allocations, but they will sometimes pick the time to do it based on their instincts and hunches. Heck, yes, this is market timing but I think it's pretty harmless.

[sscritic]

I didn't claim I could tell you. I just said that "forces you to sell high and buy low" is too simplistic. Sometimes rebalancing takes the form of "sell low but buy lower" and sometimes the form of "buy high but sell higher." Neither of these is "sell high buy low."

P.S. I am using two subjects. Do you know if sell high and buy low is referring to one subject or two?

[dandan14]

To me, the easiest way to rebalance is with new money. As I have new money to deposit, I look at my allocation and see where the money needs to go to bring things back in line. To me, these little nudges one way or the other are much easier than making big changes at the end of the year.

[Tom_T]

I didn't claim I could tell you. I just said that "forces you to sell high and buy low" is too simplistic. Sometimes rebalancing takes the form of "sell low but buy lower" and sometimes the form of "buy high but sell higher." Neither of these is "sell high buy low."

Agreed. Anyone who rebalanced from bonds into stocks in early 2009 wasn't selling high. Both stocks and bonds had fallen -- and both went on to have gains over the next four years.

[grayfox]

Did not want to hijack the thread on rebalancing after 50% market drop.

So... what's the point of rebalancing again? Especially in the accumulation stage.

To rebalance is to time the market, and since I'm certain that my ability to time the market is horrible, I just don't rebalance. But who knows, maybe the next down turn or run-up will change that thinking.

What are your thoughts?

I think your first impression is accurate. However once you drink the kool-aid and become a rebalancing drone, you will be discussing how you rebalance to controls risk and increase returns with a rebalancing bonus.

[ihckennedy]

Is your portfolio diversified into different asset classes with the objective of reducing overall portfolio risk? (This is investing 101, so I presume the answer is yes).

Well, if you don't rebalance, you undermine the whole point of diversification.

Rebalancing is the opposite of market timing: market timers adjust their asset allocation tactically in the (deluded) belief that they can predict future returns of various asset classes. To rebalance is to acknowledge ignorance of future returns and the importance of maintaining the same risk allocation you had laid out in your investment plan as being appropriate to your situation and objectives.

Research has shown that rebalancing does not necessarily result in better returns than those of a portfolio that is not rebalanced, but it does result in a lower and more consistent level of risk.

[gt4715b]

I didn't claim I could tell you. I just said that "forces you to sell high and buy low" is too simplistic. Sometimes rebalancing takes the form of "sell low but buy lower" and sometimes the form of "buy high but sell higher." Neither of these is "sell high buy low."

Agreed. Anyone who rebalanced from bonds into stocks in early 2009 wasn't selling high. Both stocks and bonds had fallen -- and both went on to have gains over the next four years.

Gov't bonds didn't fall in 2008/2009. Intermediate Treasuries were up 10% from before the crash to the market low. Even Total Bond was up about 5%.

Again, nothing magic about rebalancing and it really doesn't matter much in the early to mid accumulation stage. But as you get closer to retirement it can keep your portfolio from drifting away from where you want your portfolio.

[sscritic]

Gov't bonds didn't fall in 2008/2009. Intermediate Treasuries were up 10% from before the crash to the market low. Even Total Bond was up about 5%.

01/23/2008 $11.14
12/15/2008 $9.92

That was just one government, the one I know about.

P.S. How do you go up to a low? Sure, pick an advantageous starting point out of the past, but then that wasn't part of the decline, was it? Or are using a different market? I am sure that some bonds went up even as milk prices went lower, but if you are looking at the bond market, I don't get the "up to the low."

[paulsiu]

Yes, I do, but I am lazy and don't rebalance on a regular interval. So if the market drops and I happen to check the portfolio at the time, I will probably rebalance. I used to have very hard rule and check it every day and rebalance it immediately when it goes like 5% off, but have decided it that it was a waste of my time.

For taxable account, I often redirect purchases instead of selling especially if it will result in a big tax bill. So if stock falls, I'll alter my contribution to buy more stock.

Paul

[Bungo]

If, over the long run, stocks outperform bonds, then without adjustment, my portfolio will become more stock-heavy over time, which is the opposite of what I want as I get older and closer to retirement. That alone would justify rebalancing.

[gt4715b]

Gov't bonds didn't fall in 2008/2009. Intermediate Treasuries were up 10% from before the crash to the market low. Even Total Bond was up about 5%.

01/23/2008 $11.14
12/15/2008 $9.92

That was just one government, the one I know about.

P.S. How do you go up to a low? Sure, pick an advantageous starting point out of the past, but then that wasn't part of the decline, was it? Or are using a different market? I am sure that some bonds went up even as milk prices went lower, but if you are looking at the bond market, I don't get the "up to the low."

I was referring to the stock market low. I was saying intermediate Treasuries, as measured from sometime before the market correction to the bottom of the correction, were up around 10%, while stocks were down around 50%, thus giving one the opportunity to rebalance.

Anyway, I believe you may be looking at price data, not total return data. Here are the cumulative Total Return NAV from Jan. 2008 to Dec. 2009, i.e investing $1 on 12/31/07. Bonds, especially Treasuries did their job.

Month R3000 R3000V R2000 Total Bond Interm. Treas
12/31/2007 1.00 1.00 1.00 1.00 1.00
01/31/2008 0.94 0.96 0.93 1.02 1.03
02/29/2008 0.91 0.92 0.90 1.02 1.05
03/31/2008 0.90 0.91 0.90 1.02 1.05
04/30/2008 0.95 0.96 0.94 1.02 1.03
05/31/2008 0.97 0.96 0.98 1.01 1.02
06/30/2008 0.89 0.87 0.91 1.01 1.02
07/31/2008 0.88 0.87 0.94 1.01 1.03
08/31/2008 0.90 0.88 0.97 1.02 1.04
09/30/2008 0.81 0.82 0.90 1.01 1.05
10/31/2008 0.67 0.68 0.71 0.98 1.07
11/30/2008 0.62 0.63 0.63 1.01 1.11
12/31/2008 0.63 0.64 0.66 1.05 1.13
01/31/2009 0.57 0.56 0.59 1.04 1.12
02/28/2009 0.51 0.49 0.52 1.04 1.11
03/31/2009 0.56 0.53 0.56 1.05 1.13
04/30/2009 0.62 0.59 0.65 1.06 1.12
05/31/2009 0.65 0.62 0.67 1.07 1.11
06/30/2009 0.65 0.62 0.68 1.07 1.10
07/31/2009 0.70 0.67 0.75 1.09 1.10
08/31/2009 0.73 0.71 0.77 1.10 1.11
09/30/2009 0.76 0.73 0.81 1.11 1.12
10/31/2009 0.74 0.71 0.76 1.12 1.12
11/30/2009 0.78 0.75 0.78 1.13 1.14
12/31/2009 0.80 0.76 0.84 1.11 1.11

[allsop]

To me, the easiest way to rebalance is with new money. As I have new money to deposit, I look at my allocation and see where the money needs to go to bring things back in line. To me, these little nudges one way or the other are much easier than making big changes at the end of the year.

I do more or less the same.

Edit: Most of the savings are in taxable and I would like to avoid capital gains taxes.

[sscritic]

I believe you may be looking at price data, not total return data.

Yes, that's because when I rebalance, I have to sell my shares at a price, I don't get to sell them at a total return. The same thing holds when I buy shares, I pay a price, not a return.

[Peter Foley]

Bungo said

If, over the long run, stocks outperform bonds, then without adjustment, my portfolio will become more stock-heavy over time, which is the opposite of what I want as I get older and closer to retirement. That alone would justify rebalancing.

I agree with that reasoning. Excluding risk tolerance, it is the primary reason one should rebalance. Note that in this context rebalancing would not be done very often and could be accomplished through redirection of new deposits if rebalancing bands are fairly narrow.

[gt4715b]

I believe you may be looking at price data, not total return data.

Yes, that's because when I rebalance, I have to sell my shares at a price, I don't get to sell them at a total return. The same thing holds when I buy shares, I pay a price, not a return.

But I assume your bond fund pays interest and that you reinvest it, right? Since bonds deliver most of their return in income, looking at price data is usually misleading. You got the total return, whether you realize it or not.

[sscritic]

I believe you may be looking at price data, not total return data.

Yes, that's because when I rebalance, I have to sell my shares at a price, I don't get to sell them at a total return. The same thing holds when I buy shares, I pay a price, not a return.

But I assume your bond fund pays interest and that you reinvest it, right? Since bonds deliver most of their return in income, looking at price data is usually misleading. You got the total return, whether you realize it or not.

Yes, I see my mistake now. Rebalancing is based on percentages. If you have $100k in bonds, it doesn't matter if you got there from $100k earning no interest or $100k dropping to $50k in NAV while earning $50k in interest.

I am curious. Did you run your numbers through a rebalancing machine? Say start at 50/50 and rebalance at the end of every month. As the stock market falls, you will be buying more and more shares of stock (and watching them fall even more); when the market turns, you will be selling your shares as they go up.

I have to admit I was chicken. I didn't sell bonds to buy stocks; I took cash and bought more bonds as the price fell (I was still paying the price when I bought).

[magician]

I always find threads like this one amusing, mainly because everyone seems to assume that there is only one way to rebalance: to maintain a constant portfolio mix.

There are rebalancing strategies other than constant mix.

Then there are posts such as:

Re-balancing is the single most effective way to make your portfolio grow faster.

If by rebalancing you mean to maintain a constant mix, then it will make your portfolio grow faster (than a buy-and-hold strategy) in some scenarios, but not in others. A constant-mix strategy outperforms a buy-and-hold strategy in a volatile, non-trending market, but underperforms in a trending market. It certainly isn't universally a better strategy than buy-and-hold.

[BBL]

I always find threads like this one amusing, mainly because everyone seems to assume that there is only one way to rebalance: to maintain a constant portfolio mix.

There are rebalancing strategies other than constant mix.

An oldie that expands on that:

http://www.stanford.edu/class/msande348 ... Sharpe.pdf

For the curious.

[magician]

I always find threads like this one amusing, mainly because everyone seems to assume that there is only one way to rebalance: to maintain a constant portfolio mix.

There are rebalancing strategies other than constant mix.

An oldie that expands on that:

http://www.stanford.edu/class/msande348 ... Sharpe.pdf

For the curious.

It's funny that they describe buy-and-hold as a dynamic strategy.

Good paper, by the way. I was thinking of CPPI when I wrote the underlined sentence.

[Random Musings]

I rebalance according to my written investment plan.

RM

[BBL]

It's funny that they describe buy-and-hold as a dynamic strategy.

Good paper, by the way. I was thinking of CPPI when I wrote the underlined sentence.

I find that 'dynamic' moniker odd each time I read it. Not quite my idea of dynamic.

There aren't too many 19 year old papers I still reference but a few of them stand the test of time.

[gt4715b]

I am curious. Did you run your numbers through a rebalancing machine? Say start at 50/50 and rebalance at the end of every month. As the stock market falls, you will be buying more and more shares of stock (and watching them fall even more); when the market turns, you will be selling your shares as they go up.

I have to admit I was chicken. I didn't sell bonds to buy stocks; I took cash and bought more bonds as the price fell (I was still paying the price when I bought).

Those data are cumulative returns for iShares ETFs for Russell 3000 (IWV), Russell 3000 Value, Russell 2000, Intermediate Gov't (IEI) and Total Bond (AGG) using NAV Total Return data from the iShares website.

I just did a quick study to examine the difference between no-rebalancing and monthly rebalancing. This assumes you invest $2 on 12/31/2007. There are 3 portfolios:

P1: All stocks (Russell 3000)
P2: 50/50 Russell 3000/Intermediate Treasury mix with no rebalancing
P3: 50/50 Russell 3000/Intermediate Treasury mix with monthly rebalancing

As you can see, the monthly rebalancing during the stock market downturn causes you to in effect "double down" on the stock market and yields a maximum portfolio loss almost as great as holding all stocks! As the market recovers, the additional stocks the monthly rebalance portfolio help it recover faster and by the time the stock market reaches it's starting value, the monthly rebalance portfolio has gained more, but that's one heck of a ride especially if you wanted the risk of a 50/50 portfolio.

I think you were right not to blindly rebalance during the downturn. I don't think the effect shown is widely talked about but it's very important because it can cause people that do rebalance to freak out and sell out close to the bottom and lose a lot of money.

Month All Stocks No Rebalance Monthly Rebalance
12/31/2007 2.00 2.00 2.00
01/31/2008 1.88 1.97 1.94
02/29/2008 1.82 1.96 1.91
03/31/2008 1.81 1.96 1.91
04/30/2008 1.90 1.98 1.97
05/31/2008 1.94 1.99 1.98
06/30/2008 1.78 1.91 1.83
07/31/2008 1.76 1.91 1.83
08/31/2008 1.79 1.94 1.88
09/30/2008 1.62 1.86 1.72
10/31/2008 1.34 1.74 1.44
11/30/2008 1.23 1.73 1.38
12/31/2008 1.25 1.76 1.43
01/31/2009 1.15 1.69 1.29
02/28/2009 1.03 1.62 1.15
03/31/2009 1.12 1.69 1.27
04/30/2009 1.24 1.73 1.39
05/31/2009 1.30 1.75 1.45
06/30/2009 1.31 1.75 1.44
07/31/2009 1.41 1.80 1.56
08/31/2009 1.46 1.84 1.63
09/30/2009 1.52 1.88 1.71
10/31/2009 1.48 1.86 1.67
11/30/2009 1.56 1.92 1.80
12/31/2009 1.61 1.91 1.80
01/31/2010 1.55 1.90 1.77
02/28/2010 1.60 1.94 1.84
03/31/2010 1.70 1.98 1.94
04/30/2010 1.74 2.01 2.00
05/31/2010 1.60 1.96 1.87
06/30/2010 1.51 1.93 1.79
07/31/2010 1.61 2.00 1.94
08/31/2010 1.54 1.97 1.88
09/30/2010 1.68 2.05 2.06
10/31/2010 1.75 2.09 2.15
11/30/2010 1.76 2.09 2.15
12/31/2010 1.88 2.12 2.25
01/31/2011 1.92 2.15 2.31
02/28/2011 1.99 2.18 2.38
03/31/2011 2.00 2.18 2.39
04/30/2011 2.05 2.23 2.49

[englishgirl]

Interesting data. I used to buy in to the whole "rebalance decreases risk" thing, but then when I tried to do some math, it seemed to result in larger losses in a downturn. Which seemed like increasing risk - which I suppose it was, because if stocks are tanking, keeping the money in bonds reduces your risk. In an up market, rebalancing seems to work to keep your risk where you want it, and to take profits out by converting stock gains into bond holdings.

As it is, 50% of my portfolio is in Balanced Index. It rebalances itself daily. It makes up 85% of my 401k, and I have the 401k set to automatically rebalance every 6 months, in April and October. So, I left that rebalancing alone, and it did its thing without input from me. However, while I occasionally looked at my spreadsheet to see how close I was to my rebalance bands, I don't remember pulling the trigger and doing a manual rebalance. Or maybe I did it once. Same in my Roth IRA. If I recall correctly, I was too chicken to do it, and too chicken to see how out of whack my portfolio was getting. But I don't think it got SO bad, as it wasn't like bonds were returning 20% while the stocks were tanking.

[sscritic]

Thank you for posting these numbers. It is interesting that the no rebalance and the monthly rebalance both get back to even (2.0) at the same time and then stay roughly equal for the next five months. Rebalancing only pulls away after 9/30/10.

04/30/2010 1.74 2.01 2.00
09/30/2010 1.68 2.05 2.06

Since these are just portfolio values, I think what we are seeing is that not all 2's are equal. The rebalanced 2.0 was more stock than the no rebalanced 2.0, or so it seems.

[gt4715b]

Thank you for posting these numbers. It is interesting that the no rebalance and the monthly rebalance both get back to even (2.0) at the same time and then stay roughly equal for the next five months. Rebalancing only pulls away after 9/30/10.

04/30/2010 1.74 2.01 2.00
09/30/2010 1.68 2.05 2.06

Since these are just portfolio values, I think what we are seeing is that not all 2's are equal. The rebalanced 2.0 was more stock than the no rebalanced 2.0, or so it seems.

Right, the no-rebalanced portfolio is only 43% stocks on 4/30/2010, whereas the rebalanced portfolio is of course 50%. So the rebalanced portfolio starts to pull away as the stock market keeps recovering.

[dbr]

As a risk management tool? Grows the whole portfolio faster?

None of it makes sense. How is my portfolio less risky if I rebalance when stocks go down? Wouldn't it put more money at stake?

Risk management does not mean reducing risk. It means keeping the risk at a target. The proxy for risk is stock/bond allocation so the object is to keep no more and no less risk than desired by keeping no more and no less fraction in stocks at all times. Risk also does not mean only the chance of losing money; it also means the chance of making money. Risk, meaning the spread in the range of possible future outcomes, means greater chances of both high outcomes and low ones.

In the specific instance that stocks go down and the fraction in stocks becomes less and the size of the portfolio shrinks, then moving money to stocks to recover the original fraction does not put more money at risk because it becomes the same original fraction of a smaller total. If one did not rebalance, the money at risk would be even less.

[Rodc]

Rebalancing is over sold. Just ask Jack Bogle.

Over long periods of time (many years) it is good for helping to keep your risk in check (so stocks don't come to overwhelm your portfolio and risk budget). It is good during really wild times like 2001 or 2008-2009 (as long as the market really does snap back).

But it is just not all that most of the time. You move a few percent here or there, big deal. Shoot, no one knows their optimal stock/bond split to better than +/- 10%, maybe +/- 20%. Rebalance with bands, once a year, or every 3 years, whatever. Over the course of your life it will likely help. On the other hand if every 5 years or so you stop and reassess your plans and adjust to your changing needs, you in effect rebalance for other reasons so who cares about formal rebalancing.

In fact in general rebalancing reduces risk and reduces returns because more often than not you sell high return assets to buy low return assets (though not always).

For a long time in accumulation you can keep balanced enough just by directing new money to any asset that is below target, and if something like 2008-2009 comes along you can either just keep pumping new money into stocks or you can sell bonds to buy stocks (my choice but for some that was too scary).

I rebalance, but I don't expect much out of it, not really a big deal.

[magician]

As a risk management tool? Grows the whole portfolio faster?

None of it makes sense. How is my portfolio less risky if I rebalance when stocks go down? Wouldn't it put more money at stake?

Risk management does not mean reducing risk. It means keeping the risk at a target. The proxy for risk is stock/bond allocation . . . .

One's stock/bond allocation is not the proxy for risk. Some people (perhaps many people) use it as a proxy for risk, but there are certainly others. Calling it the proxy for risk suggests that it is the only one; it isn't. It's arguably not even the best one.

Risk also does not mean only the chance of losing money; it also means the chance of making money. Risk, meaning the spread in the range of possible future outcomes, means greater chances of both high outcomes and low ones.

To some people, that's exactly what risk means. If you ask 20 people what "risk" is, you'll get 20 different answers; who's to say which one is correct? I submit that the correct answer depends on the individual, and certainly for some individuals the possibility of loss is exactly what risk is to them.

In the specific instance that stocks go down and the fraction in stocks becomes less and the size of the portfolio shrinks, then moving money to stocks to recover the original fraction does not put more money at risk because it becomes the same original fraction of a smaller total.

If you don't rebalance, you have, say, $50 in stocks, so you have $50 at risk (of loss). If you rebalance, you have, say, $60 in stocks, so you have $60 at risk (of loss). Sixty dollars is more than fifty dollars, so you have more money at risk.

Furthermore, if stocks keep going down, a constant-mix rebalancing approach has a floor portfolio value of $0, whereas a buy-and-hold approach has a floor value equal to your (presumably non-zero) bond allocation.

[dbr]

New resolution effective today: Do not participate in any discussions about risk or about rebalancing.

Proposed resolution for tomorrow: Stop posting to this forum.

[Garco]

Normally I do a major rebalance at the end of the year. That's also when I make a general assessment of my portfolio and reset my AA. My basic allocation has been moving in a more conservative direction for the past few years, and that tends to dominate other strategic reallocations or tactical rebalancing.

During the year if there's been a serious runup in one sector or fund, I consider taking some money off the table. At the end of the first quarter of 2012, for example, my AA had shifted about 4 points toward equities since the beginning of the year, and I brought things into line at that time. This had the effect of locking in some very substantial gains. I'm kind of on the lookout for this to happen again this year, and would likely sweep some money out of equities if things continue as they have. It's not hard for me to decide to do this because I'm basically reinforcing the multiyear trend to derisking my investments.

[telemark]

If this is "market timing," it is very different from what people who call themselves "market timers" do. It does not involve large-commitment shifts in an out of asset classes, just small adjustments, and it isn't based on technical analysis.

I would define market timing in general as taking actions based on what you think is going to happen in the market in the future. Rebalancing makes no predictions about the future: it's based solely on the state of your portfolio today.

[telemark]

None of it makes sense. How is my portfolio less risky if I rebalance when stocks go down? Wouldn't it put more money at stake?
TP

In this case your portfolio has become more conservative, and rebalancing restores it to the level of risk you had previously chosen. This is a good thing: when stocks start to go up again you want to have some money in them. Sometimes rebalancing lowers risk, and sometimes it increases risk. The same holds true for returns.

[Toons]

Gradual rebalance with new money or dividends and cap gains.Keep it simple

[camontgo]

Risk also does not mean only the chance of losing money; it also means the chance of making money. Risk, meaning the spread in the range of possible future outcomes, means greater chances of both high outcomes and low ones.

To some people, that's exactly what risk means. If you ask 20 people what "risk" is, you'll get 20 different answers; who's to say which one is correct? I submit that the correct answer depends on the individual, and certainly for some individuals the possibility of loss is exactly what risk is to them.

Definitions of risk may differ from one individual to another. However, many of us have definitions of risk which, on closer examination, are not consistent with the way we actually behave when faced with uncertainty.

For example, I used to believe that I only cared about “downside risk”, but then someone explained that if I was truly risk-neutral on the upside then I would be indifferent between a wildly uncertain upside bet (say a coin flip for $1 million or $0) and getting the average value for certain. That's not my true preference. After thinking through a few more realistic scenarios, I decided I did care about that upside risk after all.

It is true that smart people will probably always disagree on a definition of risk, but simply going with “who's to say what's correct” cuts things off before the thinking even gets started. There may not be a universal definition out there, but there is a lot of sloppy thinking about risk (probably including some of my own). Debating definitions of risk helps reduce the latter even if we never find the former.

[scrabbler1]

If, over the long run, stocks outperform bonds, then without adjustment, my portfolio will become more stock-heavy over time, which is the opposite of what I want as I get older and closer to retirement. That alone would justify rebalancing.

I agree. As an early retiree, most of my rebalancing is in my IRA so there are no tax consequences. After years of 55/45 in favor of stocks, I am moving to a 50/50 AA, selling off some of its stock fund holdings to buy more in its bond fund holdings. I do some rebalancing in my taxable account but I have a wider AA range because my taxable accounts have as its main purpose to generate enough monthly dividends to cover my expenses while keeping some stock funds around as a growth pieve.

[magician]

It is true that smart people will probably always disagree on a definition of risk, but simply going with “who's to say what's correct” cuts things off before the thinking even gets started.

I disagree.

If there were one, universally accepted definition of risk, then all discussion would end with the presentation of that definition. When there is disagreement about what is correct, then the opportunity is ripe for discussion about which potential definitions have merit, which don't, which apply to some people, which apply to others, and so on.

Definitions of risk may differ from one individual to another. However, many of us have definitions of risk which, on closer examination, are not consistent with the way we actually behave when faced with uncertainty.

For example, I used to believe that I only cared about “downside risk”, but then someone explained that if I was truly risk-neutral on the upside then I would be indifferent between a wildly uncertain upside bet (say a coin flip for $1 million or $0) and getting the average value for certain. That's not my true preference. After thinking through a few more realistic scenarios, I decided I did care about that upside risk after all.

And that's an eye-opening revelation, and kudos to you for taking the time to examine your beliefs and your behaviors.

I've taught an on-line course in asset allocation in which we discussed risk at length. One of the most interesting discussions we would have in those classes arose from my asking the students, "What is risk? (That is, when you describe an investment as "risky", what do you really mean?") Most often the discussion would start with short, stock answers (e.g., standard deviation of returns, and the like), but became a lot more fruitful as I pressed the students for more detail, and for examples - such as yours - of how their definition molded their behavior. Not a single student has ended up with the same (personal) definition of risk as the one they first gave in response to that question.

[baw703916]

If this is "market timing," it is very different from what people who call themselves "market timers" do. It does not involve large-commitment shifts in an out of asset classes, just small adjustments, and it isn't based on technical analysis.

I would define market timing in general as taking actions based on what you think is going to happen in the market in the future. Rebalancing makes no predictions about the future: it's based solely on the state of your portfolio today.

I invest in stocks because I think the market will go up in value in the next 2-3 decades. If I didn't think this, then I can't imagine why I would own stocks. According to the definition, that constitutes market timing.