Skip to main content
Home Working together to build your tomorrow

The editors of the New York Times' Sunday business section must have been reading this column for years. They finally deigned to write a column endorsing my constant drumbeat on the virtues of diversifying and rebalancing -- creating the "path of minimum regret," we call it.

On May 6, Alex Tarquinio wrote what I consider a definitive work on the subject. He pointed out that a mix of Vanguard's 500 index, small cap index and international index (in a 60 percent-20 percent-20 percent mix) would have matched the March 2000 market high as early as November 2004. The 500 index by itself would have taken two years longer, until October 2006.

In my old booklet "Roadmap to Riches," updated every few years, I have graphed the comparative results of (you guessed it) the 500 index, the small cap index and the foreign fund index. The path of minimum regret is a composite result of an equal investment in each fund.

Investing equal amounts in these funds was just the start. Rebalancing once a year equalized the amount of money in each fund, taking chips off the table from winners and adding to losers.

Why was this important? Because without rebalancing, and just letting sleeping dogs lie, the account balances for each investment developed, widely varying the amounts of money.

The temptation to take from a loser and add to a winner under those circumstances is overwhelming for most of us. It explains why the average mutual fund investor in the 1980s and '90s averaged only a 3 percent annual return, while the average fund was earning 16 percent per year.

On a spreadsheet, I assume an annual 401(k) investment of $10,000 beginning in 1988, deposited in equal amounts into the three investments. An unbalanced account would have had $137,000 in the index, $98,000 in the small cap and $81,000 in the foreign fund.

Any normal person about that time would have been slapping the side of their head and asking why they hadn't put all their money in the 500 index. It was worth $60,000 more than the foreign fund after just 11 years.

Taking that action, however, spells disaster.

By comparison, the results of rebalancing would have generated $103,000 in the 500 index, $105,000 in the small cap and $110,000 in the index. Certainly less temptation to make a dumb move.

It just so happens that by Dec. 31, 2006, both approaches (rebalanced versus the sleeping dogs) had very little difference in total value -- $547,000 for balanced and $542,000 for unbalanced.

However, this similarity of final results can be a result of two things.

First, there is quite a bit of correlation between these three investment types. They are all invested in common stocks as opposed to bonds or other investment types such as real estate.

Second, we just happen to be choosing a moment in which the planets happen to have lined up and produced similar results. The unbalanced approach just got lucky.

In more elaborate models, using REITs (real estate investment trusts) and bonds, the rebalancing can create more dramatic results.

My best example illustrates how rebalancing between a tech fund and a S&P 500 index can generate a $400,000 advantage in 30 years on the same $10,000-per-year investment.

The New York Times study, using a rebalance between a 50-50 mix of stock and bond index funds, would have increased returns by 10 percent, i.e., a 5 percent return would have increased to 5.5 percent.
Rebalancing works most effectively when the types of investments used have the most widely differing performance characteristics.

"Inverse correlation" is what oils the workings of a rebalance strategy. In the meantime, the magic of compound interest makes an additional 1 percent of annual earnings balloon our ultimate nest egg size by as much as 20 percent.

Taking steps to create what seem like barely meaningful annual results today can yield huge benefits over thirty years. Want to retire five years ahead of schedule? Rebalancing will improve your odds.

Get weekly articles delivered to your inbox!

* indicates required