John Bogle Misses the Boat on Rebalancing

2013-10-11

by Eric D. Nelson, CFA

A recent article in Money Magazine discussed the topic of portfolio rebalancing, a subject we covered a few weeks ago in Servo Thoughts.  Our conclusion was that rebalancing is a key ingredient to maintain the overall risk profile of a portfolio, and may even lead to higher returns through its sell high/buy low tendency.

In truth, our pro-rebalancing view is not a breakthrough.  Just about everyone agrees that periodically restoring an allocation to its target weights is an important portfolio management technique.  If not, what’s the point of having target weights in the first place?  Well, almost everyone.  Vanguard founder John Bogle was quoted in the Money article as saying “it’s better not to rebalance.”  Say what?  Actually, the full comment was bracketed with “If you can ignore market fluctuations (it’s better not to rebalance), since you’re likely to get higher returns.”

 

Where does Bogle miss the boat?  Of course a portfolio that becomes much riskier over time has higher returns!  This is like saying an 80% to 100% stock portfolio (remainder in bonds) is superior to 60% in stocks because you’re likely to get higher returns—true, but that comes at the expense of higher volatility, something that was unacceptable given the initial portfolio allocation’s more conservative mandate.  

 

Specifically, by not rebalancing periodically and selling stocks to buy bonds (which is what will happen more often than selling stocks to buy bonds), you eventually will have all equity or at least an equity-dominated portfolio.  But if that’s what you wanted, why not start that way in the first place?  Presumably it’s because you didn’t need the higher potential returns and didn’t want to put up with the higher volatility.  So if it didn’t make sense at first, why was it eventually acceptable?  Bogle really boggles the mind with this line of reasoning.  If the raw pursuit of higher returns were all that mattered, we’d all have 100% of our portfolios in small cap value stocks—the highest risk and expected return part of the global market.

 

Table 1: The Rebalancing Effect on Balanced Portfolios (1928-2012)

Portfolio

Annualized Return

Average Decline

00-02 Loss

2008 Loss

65/35 S&P 500 Mix

+8.6%

-8.6%

-15.7%

-19.5%

65/35 S&P 500 Mix (never rebalanced)

+9.1%

-12.1%

-37.0%

-36.4%

80/20 S&P 500 Mix

+9.1%

-10.7%

-25.6%

-27.0%

 

 

 

 

 

65/35 Asset Class Mix

+10.3%

-9.0%

-6.7% (’02)

-22.5%

65/35 Asset Class Mix (never rebalanced)

+11.4%

-14.3%

-14.7% (’02)

-42.1%

80/20 Asset Class Mix

+11.7%

-14.4%

-9.4% (’02)

-31.6%

 

Table 1 looks at the impact of not rebalancing on balanced portfolios.  The first example uses a 65% S&P 500, 35% bond portfolio and tracks its long-term returns when rebalanced annually and never rebalanced.  The return difference is 0.5% per year higher for the never-rebalanced version, but that came at the expense of much greater risk.  In particular, the average negative-year return for the rebalanced portfolio was just -8.6% vs. -12.1% for the non-rebalanced mix.  During the bear market years of 2000-2002, the never-rebalanced mix lost 21.3% and 26.9% more than the mix that was annually rebalanced.  

 

If the higher returns from the never-rebalanced portfolio were desirable, one could have simply opted for an 80% S&P 500, 20% bond portfolio for the entire period and rebalanced that annually (similar annual standard deviation as the 65/35 never-rebalanced portfolio).  This mix would have produced the same +9.1% per year return over the entire period, yet only produced an average loss of -10.7% during negative years, and declined over 11% and 9% less than the non-rebalanced portfolio in 2000-2002 and 2008 respectively.

 

Table 1 also looks at it from the standpoint of a more diversified asset class allocation.  The difference between the 65/35 annually rebalanced and never-rebalanced asset class mixes was more dramatic—+1.1% per year in favor of the never-rebalanced allocation.  This is due to the fact that over time, stocks in general and small cap value stocks (26% of the original 65/35 mix) in particular will come to dominate a non-rebalanced portfolio.  But this too comes with additional risk.  The never-rebalanced mix lost an average of -14.3% during all downturns, vs. just -9.0% for the annually rebalanced mix.  In 2002 and 2008, the differences in losses were 8% and 20% less for the rebalanced mix.  

 

Once again, if the higher returns of the non-rebalanced mix are called for, make that the policy from the start.  An 80/20 asset class portfolio (similar annual standard deviation as the 65/35 never-rebalanced portfolio) that holds a greater allocation to small value stocks produced a +0.3% per year higher return over the entire period, but did so with similar long-term average declines (-14.3%) and much less recent losses.  In 2002 and 2008, the 80/20 rebalanced portfolio lost 5% and 12% less than the non-rebalanced 65/35 version.

 

So in some cases, such as the S&P example, we find that failing to rebalance produces higher risk for a given level of achievable return.  Looking at more diversified asset class portfolios, we find that failing to rebalance produces higher risk and lower returns—exactly the opposite of what well-intentioned portfolio management hopes to achieve.

 

Why is it, then, that John Bogle is willing to accept higher risk, lower returns, or both?  It could be that he just doesn’t understand the real reason to rebalance (thinking it’s a return-enhancement strategy and not a risk-management effort), or he might not have the right data.  But most likely it’s because rebalancing is hard.  Selling what has done well to do what has done poorly goes against what feels natural and like the right thing to do.  Underperforming assets (that must be purchased periodically) are usually underperforming for a reason that prevents us from wanting to add to them.  Outperforming assets (that must be pared back periodically) have momentum, they are popular, and they feel good to own.  So this boils down to behavior.  John Bogle, who himself says he doesn’t rebalance his portfolio, is probably unable to overcome the emotional barrier to sell higher and buy lower.  Hard to imagine for someone who has spent his life in finance, but experience and emotional fortitude aren’t always correlated.  

 

And, of course, most individual investors suffer from the same emotional paralysis.  Fortunately, aligning yourself with a highly disciplined investment advisor who will ensure you continually make the proper investment decisions and not emotional-based ones is a viable alternative to navigating investment markets alone.  John Bogle may have missed the boat on this, but it’s not too late for you to come aboard.

___________________________________

 

Source of data: DFA Returns 2.0

 

65/35 S&P 500 Mix = 65% S&P 500, 35% 5YR T-Notes, rebalanced annually.

80/20 S&P 500 Mix = 80% S&P 500, 20% 5YR T-Notes, rebalanced annually.

 

65/35 Asset Class Mix = 19.5% S&P 500, 19.5% DFA US Large Value Index, 26% DFA US Small Value Index, 35% 5YR T-Notes, rebalanced annually.

80/20 Asset Class Mix = 4% S&P 500, 20% DFA US Large Value Index, 56% DFA US Small Value Index, 20% 5YR T-Notes, rebalanced annually.

 

Past performance is not a guarantee of future results.  The returns and other characteristics of the allocation mixes contained in this article are based on model/back-tested simulations to demonstrate broad economic principles.  They were achieved with the benefit of hindsight and do not represent actual investment performance. There are limitations inherent in model performance; it does not reflect trading in actual accounts and may not reflect the impact that economic and market factors may have had on an advisor’s decision-making if the advisor were managing actual client money. Model performance is hypothetical and is for illustrative purposes only.  Model performance shown includes reinvestment of dividends and other earnings but does not reflect the deduction of investment advisory fees or other expenses except where noted. Clients’ investment returns would be reduced by the advisory fees and other expenses they would incur in the management of their accounts.  Past performance is not a guarantee of future results, and there is always the risk that an investor may lose money.  Indexes are not available for direct investment.