How Often Should You Rebalance?

2/3/2014


By most measurements, 2013 was an excellent year for stock investors, with the S&P 500 Index delivering a total return of 32.4% and 94% of the index's components managing positive returns for the year. Nearly three-fourths of the stocks in the large-cap index returned more than 20%.

However, many investors have fallen victim to their own success and might need to rebalance portfolios. Stocks probably make up a larger portion of your portfolio relative to bonds than they did a year ago. And within your stock holdings, top performers account for a bigger piece of the pie, while those that lag have shrunk to lower weightings. To rebalance, you sell a portion of your winners and purchase the losers, reverting to a prior weighting.

For example: Suppose you'd started the year with your $100,000 portfolio divided between two exchange-traded funds, $50,000 in Vanguard S&P 500 ($164; VOO) and another $50,000 in Vanguard Total Bond Market ($81; BND). Assuming reinvestment of dividends, at the end of the year your portfolio would have grown to $115,000, with just over $66,000 in the stock fund and just under $49,000 in the bond fund, which lost 2% including dividends.

After stocks' strong performance, they'd make up 57.5% of the portfolio. To rebalance back to the 50/50 weighting, you'd sell enough of the stock fund to shrink the position to $57,500 and invest the proceeds in the bond fund.

By sticking with a higher stock allocation, in effect you're making a bet that stocks will once again outperform. However, according to Morningstar, long-term government bonds have topped the return of large-company stocks in 10 of the last 25 years. The Forecasts expects stocks to keep outperforming bonds over the next year, but we're not sufficiently confident to advise readers to alter their long-term asset allocation.

If you have targeted a certain mix of stocks, bonds, and other investments, rebalancing periodically should keep your portfolio's risk fairly steady over time. Of course, the key word is "periodically," because every time a stock makes a move, your portfolio deviates slightly from its target weightings. So how often should you rebalance your holdings?

Many academics and finance professionals have studied rebalancing, and while not every study yields the same results, we've identified two common trends:

• Frequent rebalancing (monthly or quarterly, or using tight tolerance bands) ramps up costs quickly because of the volume of transactions needed. Any outperformance caused by the rebalancing is often gobbled up by trading costs.

• The most effective rebalancing strategies are those that keep costs low and limit volatility. Rebalancing costs include brokerage commissions when you buy or sell and potential tax liability when you sell a security that has increased in value since you purchased it.

We have long advised readers to limit transaction costs by using a discount broker, but taxes are tougher to control. Assuming the transaction costs are reasonable, investors can afford to rebalance tax-protected accounts such as IRAs more often than taxable accounts. However, frequent rebalancing doesn't seem to pay off for investors, even without taking taxes into account.

The table below shows the results of a study by Washington Trust Bank suggesting that rebalancing no more often than quarterly or when an asset class deviates from its target weighting by more than 10% boosted portfolio returns from 1978 through 2009. The portfolio in question contained 13 mutual funds, mostly stocks and bonds with some alternative assets thrown in.

REBALANCING STRATEGIES COMPARED

Washington Trust Bank studied the effects of different rebalancing strategies from September 1978 through September 2009. The company's portfolio included five fixed-income funds, five stock funds, and three funds focused on alternative assets.

Strategies that involved frequent rebalancing (monthly or using narrow tolerance bands) lagged the 10.43% annualized return of the nonrebalanced portfolio, then tacked on high trading costs. The widest tolerance bands (30% and 40%) yielded the most volatile returns, as measured by standard deviation. Semiannual rebalancing and a 25% or 30% tolerance band (presented in bold) appear to offer the highest returns net of trading costs.

Rebalancing Strategy
Annual.
Return
(%)
Est.
Trading
Costs
(%)
Return
Net Of
Trading
Costs
(%)
Excess
Return
Over Non-
Rebalanced
Portfolio
(%)
Number
Of Trades
Standard
Deviation
Of Annual
Returns
(%)
Time periods
Monthly
10.38
0.25
10.13
(0.30)
370
10.62
Quarterly
10.46
0.15
10.31
(0.12)
124
10.58
Semiannually
10.59
0.12
10.47
0.04
62
10.54
Annually
10.49
0.09
10.40
(0.03)
31
10.63
Tolerance bands
1% Away From Target Weight
10.39
0.25
10.14
(0.29)
367
10.62
2%
10.39
0.21
10.18
(0.25)
345
10.62
10%
10.54
0.11
10.43
0.00
144
10.62
25%
10.51
0.04
10.47
0.04
39
10.65
30%
10.51
0.05
10.46
0.03
28
10.74
40%
10.44
0.03
10.41
(0.02)
13
10.89
Source: Washington Trust Bank.

Using an annual schedule or tolerance bands of more than 25% also increased portfolio volatility relative to tighter strategies. Here's how tolerance bands work:

Suppose you target a 40% weighting in stocks and use a 25% tolerance band. Since 25% of 40% is 10%, you'd rebalance when your stock weighting rose to 50% (40% plus 10%) or fell to 30% (40% minus 10%). Using that strategy, you wouldn't rebalance the two-fund portfolio introduced in the first part of this story, despite last year's unbalanced returns.

The study found strategies that rebalance semiannually or when an asset class deviates from its target by 25% or 30% provided the best returns net of trading costs, with moderate risk.

Since your trading costs could prove higher than the study's estimate, the use of a 25% or 30% tolerance band may make the most sense. In the Washington Trust study, that strategy required fewer transactions than semiannual rebalancing. Investors trading in taxable accounts should also favor the strategy that calls for the smallest number of buys and sells.


Bonds take the edge off stock portfolios

From 1926 through 2012, a portfolio of 30% large-company stocks and 70% long-term government bonds delivered positive returns in all 83 rolling five-year periods. While stock-only portfolios delivered superior long-term returns, blends of stocks and bonds tended to provide a higher percentage of up years, as well as substantially lower risk, as measured by standard deviation (the dispersion of annual returns around the long-run average).

5-Year Periods
-------- (1926 To 2012) --------
---------- 1-Year Periods (1926 To 2012) ----------
Portfolio
Avg.
Return
(Annual.)
(%)
Number Of
Positive
Returns
% Of
Positive
Returns
Avg.
Return
(Annual.)
(%)
Number Of
Positive
Returns
% Of
Positive
Returns
Standard
Deviation
Of Annual
Returns
(%)
Return/
Risk
Ratio
100% large-
company stocks
9.8
71
86
11.8
63
72
20
0.58
70% stocks /
30% bonds
8.9
78
94
10.0
66
76
15
0.69
50% stocks /
50% bonds
8.2
78
94
8.9
68
78
11
0.79
30% stocks /
70% bonds
7.3
83
100
7.8
70
80
9
0.85
100% long-term
gov't bonds
5.7
77
93
6.1
65
75
10
0.63
Source: Morningstar.

Use bonds to control risk

Periods when one asset class strongly outperforms another can cause investors to forget about diversification. After all, while we diversify portfolios to protect against volatility, we really only worry about downside volatility.

2013 provided an awesome display of upward volatility for stocks, but investors banking on a repeat are making a mistake.

Could the S&P 500 Index return another 32%? Certainly. Or it could post a 20% decline and lag the bond market. Neither of those extremes seems likely, but history suggests a healthy mix of stocks and bonds offers investors a better blend of risk and return.

From 1926 through 2012, a portfolio of 30% large-company stocks and 70% long-term government bonds delivered positive returns in all 83 rolling five-year periods. While stock-only portfolios delivered superior long-term returns, blends of stocks and bonds tended to provide a higher percentage of up years, as well as substantially lower risk, as measured by standard deviation (the dispersion of annual returns around the long-run average).

Adding bonds to a stock portfolio decreases the average return but boosts return per unit of risk, as shown in the table above.


Current Hotline

Stock Spotlight

Individual Stock Reports

ISRs make stock research easy!

Perhaps the most valuable two page reports available anywhere.

All the data you would normally have to plow through years of 10-K filings, earnings reports, and reams of market data to assemble — yours all in one concise report.

ISRs contain our proprietary Quadrix scores — find out how we rate all the stocks in the S&P 500.

Visit us at individualstockreports.com