I feel like my last post about rebalancing wasn’t as thorough as I’d have liked it to be, so here I go again, adding some quick definitions and including a review of several research articles about the subject.
What is Rebalancing?
Let say you examine your risk tolerance and decide to invest in a mixture of 70% stocks and 30% bonds. As the years go by, your portfolio will drift one way or another. You may drop down to 60% stocks or rise up to 90% stocks. The act of rebalancing involves selling or buying shares in order to return to your initial stock/bond ratio of 70%/30%.
Why Rebalance?
Rebalancing is a way to maintain the risk/reward ratio that you have chosen for your investments. In the example above, doing nothing may leave you with a 90% stock/10% bond portfolio, which is much more aggressive than your initial 70%/30% stock/bond mix.
In addition, rebalancing also forces you to buy temporarily under-performing assets and sell over-performing assets (buy low, sell high). This is the exact opposite behavior of what is shown by many investors, which is to buy in when something is hot and over-performing, only to sell when the same investment becomes out of style (buy high, sell low).
However, in taxable accounts, rebalancing will create capital gains/losses and therefore tax consequences. In some brokerage accounts, rebalancing will incur commission costs or trading fees. This is why, if possible, it is a good idea to redirect any new investment deposits in order to try and maintain your target ratios.
How Often Should I Rebalance My Portfolio?
Some people rebalance on a certain time-based schedule – for example, once every 6-months, every year, or every 2 years. Others wait until certain asset classes shift a certain amount away from their desired targets before taking any action. A good source of research articles about which method is optimal can be found at the AltruistFA Reading Room. I’ve been reading through them the past few days, and I’ll try to provide a very general overview of the articles here.
So what is best? You may be surprised by the fact that not only is there no clear agreement on the answer to this question, but many of the articles actually contradict each other! For instance, compare this Journal of Investing article:
Over this period, regular monthly rebalancing returns dominated less active approaches. Should one infer that daily rebalancing is better still? Our data cannot say, but it seems plausible.
with this excerpt from an Efficient Frontier article:
So, what can we conclude from all this? Monthly rebalancing is too frequent. There are small rewards to increasing one’s rebalancing frequency from quarterly up to several years, but this comes at the price of increased portfolio risk.
Eh? I believe that this is because their results vary significantly with the time period chosen and asset classes being used in their back-tested scenarios.
Then there is this paper from Financial Planning magazine, which used the 25 year period from Oct. 1977-Sept. 2002 and a 60% Stock (S&P 500 Index) and 40% Bond (Lehman Bros. Gov’t Index) as the starting/target allocation. Here are the results for various rebalancing frequencies:
![altext](https://www.mymoneyblog.com/images/0707/rebal1.gif)
The various rebalancing periods showed minimal performance differences, although annual rebalancing held a slight return margin and a higher risk margin.
Because the risk-adjusted performance differences among the portfolios were small, the answer to the question of when to rebalance–monthly, quarterly, semi-annually, or annually–depends mainly on the costs to the investor of rebalancing.
Efficient Frontier’s Bernstein also agreed in the this last respect, stating “The returns differences among various rebalancing strategies are quite small in the long run.”
In the “wait for a significant shift before taking action” camp is author Larry Swedroe, who I think also presents a very reasonable solution. From a WSJ article:
With major holdings like U.S. stocks, foreign stocks and high-quality U.S. bonds, consider rebalancing whenever your fund holdings get five percentage points above or below your targets, suggests Larry Swedroe, research director at Buckingham Asset Management in St. Louis. For instance, if you have 40% earmarked for bonds, you would rebalance if your bonds got above 45% or fell below 35%.
Meanwhile, for smaller positions in sectors like emerging markets and real-estate investment trusts, Mr. Swedroe recommends a 25% trigger. So if you have 5% targeted for emerging-market stocks, you’d rebalance if emerging markets balloon above 6.25% or fall below 3.75%. “You definitely want to be rebalancing, but you don’t want to be doing it too often,” Mr. Swedroe says. “You want to let stocks go up a bit before you sell, but not so much that you lose control of risk.”
Summary
Since it seems that there is no concrete right answer, I think the most important thing is to just make sure you set up some way to rebalance that does not involve any emotions or market timing. Don’t worry about the details, but don’t let your portfolio run off on its own either. I think the subtitle of one of the articles above sums it up quite well… ‘Tis Better To Have Rebalanced Regularly Than Not At All.
I have personally chosen to rebalance annually. This method keeps it simple while still controlling risk and offering potential extra return. If I recall correctly, it is also recommended in Ferri’s book All About Asset Allocation (review).
I very much agree with you that rebalancing should not be triggered by market timing. Even when a certain asset class has appreciated significantly, you still don’t know if it has hit its peak. Two years ago, we were all told that emerging markets were way over-valued. I considered reducing my exposure, but didn’t because of the taxes I would pay. Since then, emerging markets have continued to go up at about 30% a year. I try to rebalance through contributions, rather than selling funds–particularly if the fund is in a taxable account.
I think this is why I prefer the lifestyle/target fund option for my 401k and IRA, which allocates according to an anticipated retirement date, in my case 2035. The allocation adjusts automatically each quarter. Less freedom of choice, but also a no brainer.
Why is the standard deviation in the above table so high?
the chart didn’t take into consideration tax liabilities for more frequent rebalancing.
second, i’m one that feels rebalancing should be done based off of targets/goals you have set vice specific time periods. i would also contend that variations in answers occur dependent upon the makeup of a portfolio. If you are in a very aggressive portfolio, you might opt to rebalance things sooner than later vice someone who is in a more balance portfolio. I don’t think you can compare apples to oranges in this case. again, you need to have goals for your investments and safety triggers in case of market down turns. even though you are saving long term, doesn’t mean that you shouldn’t be actively engaged in your investments.
fyi, Tommy Vu is now playing poker. Lol, I first saw the informercial on your website (via Youtube):
link
How did they calculate the Sharpe Ratio? It should be excess returns divided by standard deviation, but the numbers don’t work out in that table. Sharpe ratios in excess of 1 would imply excess returns greater than the standard deviation, which is not the case with this data.
Nomad, the Sharpe Ratio calculation does look funky.
Onto the topic at hand… The best thing about max-mean min-variance portfolio optimization is that it gives you one unique point where you max returns per unit of risk. So without needing to know a lot of calc and in a world without transaction costs you should be rebalancing your portfolio everytime your asset class allocation are not optimized (i.e. you rebalance your portfolio infinitely with very infinite small market movement) assuming returns, correlation, variance, risk free rate, and risk tolerance are held constant.
A useful and simple portfolio performance calculation is your policy index. Its your original allocation (i.e 60% stock/40% bonds ratio held constantly through time) multiplied by the original funds that you were in. Compare this to how you actually perform tells you if you gained or lost by not rebalancing.
The trend these days are that many investment managers are producing more global tactical asset allocation products (GTAA or GAA for short). There are different types and strategies of GAA funds but one type, “Opportunistic” funds, take bets on all different types of beta around the world so they essentially rebalance for you with their bets.
Regards,
BOM
I’ve set up an Excel spreadsheet that tells me when to rebalance per Larry Swedroe’s recommendation.
One sheet is set up to do an automatically updating web query from yahoo finance of my entire portfolio.
The prices from that sheet update the numbers on the main sheet automatically every time I open the file.
I use the conditional formatting feature on the row for each fund on the main sheet to highlight it as red if it’s gone below my desired allocation or blue if it’s gone above my desired allocation. For funds where yearly limits prevent me from rebalancing, such as the funds in my Roth, I’ve skipped the conditional formatting.
I put all my retirement money into the Vanguard 2045 Target Retirement fund. So, someone does the rebalancing for me.
But having said that, I think regular rebalancing is good. If nothing else, it forces you to take a look where you are.
Nice article. Love your blog.
Regarding Vanguard 2045…. do they actually rebalance the way above? Or are they going to just sell off whatever’s high to pay their fees? I read their prospectus and didn’t see a clear answer to that. Also, the 2045 fund is 5% bonds and 95% stocks right now… not necessarily the best stragety even for a young 30 something.
If you think about it, all Target Funds essentially balance daily. Any money coming into these funds will be used to purchase asset classes that haven’t performed as well as the others to keep the target allocations in line. If a withdrawal is made, I would imagine it would be taken from the best performing asset class to bring that percentage back into line. These funds have such large amount of money moving around that I doubt they ever need to choose a time to “rebalance” because they do it through natural asset flows.
I’m guessing these portfolios will outperform over time. While most people think 5% in bonds is too much, keep in mind if stocks tank, the bond money will provide the powder to buy stocks on the cheap. 5% isn’t too bad of an allocation.
I don’t rebalance monthly but I do check my investments in a spreadsheet I’ve set up to track them.
After completing my review I determine whether I need to rebalance or not. Given the speed at which markets move today I find checking up on things once a month to be beneficial in maintaining discipline while investing.
Bernstein posted up his latest article:http://www.efficientfrontier.com/ef/0adhoc/comin.htm
and he uses the dreaded “bubble” word. Anyone else think this is a “bubble”?
Myself, I wouldn’t call it a bubble but my gut tells me we are in for correction at some point. This really doesn’t matter for us “asset allocation” people because we know market timing doesn’t work. However we may be tempted by the siren call of tinkering and adding some whack little ETF’s, as Bernstein mentions.
Ted says, “…Also, the 2045 fund is 5% bonds and 95% stocks right now? not necessarily the best stragety even for a young 30 something.”
No it’s not it is 10.2% bonds, 0.19% short-term reserves, and 89.79% stocks. link
I am 33 and have been investing for 7 years now with multiple decades of investing in ahead of me. I never rebalanced my portfolio. And hopefully never will.
Why? ’cause I don’t want to cut my winners and add to my losers! In my book that is what a re-balancing does. Instead I evaluate my individual positions for what they are and make a buy/sell decision based on that not on what percent they make up in my portfolio.
NotManyOptions….
Four words: “Regression to the Mean”
Are you still letting those winning tech stocks run? I don’t suppose you would have added to those “loser” international stocks back in the early 2000s, after all, why add to losers!
I trust regression to the mean much more than I trust anyone’s ability to hop in and out of positions at the right time… It hasn’t failed in 80 years.
I found the same results on when to rebalance in a couple books by Siegel. Annually works best for me, for a host of reasons.
I rebalance and then dont even look at the market, it is very freeing.
Takes about 20 minutes door to door, annually. I just use a program called the portfolio rebalancer and my account statements. was only 4 trades required this year.
davis Says: where can I find the program”portfolio rebalancer”?
Thanks…
Tactical based rebalancing is the best way to go about. Periodic rebalancing is also an option, but no rebalancing ever guarantees the best result,