Seeking Alpha

How Often Should You Rebalance A Portfolio?

by: Long/Short Investments
Long/Short Investments
Long/short equity, macro, hedge fund analyst
Summary

This article is dedicated to those who invest mostly hands-off.

So you’ve decided you want to invest passively, figured out what assets to buy, what allocations to hold them in, how much leverage (if any).

Now it’s a matter of determining how often you need to rebalance a portfolio, or keep allocations in line with your goals and preferences.

Do you need to stay on top of things monthly? Quarterly? Semi-annually? Annually? Never?

I use backtesting on a sample mix of assets to suggest an answer to the question.

Long-term passive investing is one of the best strategies to take, especially among individual investors who are simply trying to secure a stronger financial future at minimal cost and headache.

Even institutional investors, in my opinion, should probably cut down on their overall level of activity inherent in their strategies. Most have too much confidence in their ability to predict the future and don't generate alpha, or risk-adjusted returns in excess of the broader market. There is no real evidence that more activity leads to more gains.

The playing field between retail and institutional investors has already been heavily leveled over the past few decades, and sound passive investing over a long enough time horizon can just about whittle away the advantages entirely. It's my opinion that individual investors should focus on having a balanced asset allocation that balances risk between assets to help best linearize gains over time.

Trying to read the global headlines for signs of what to do with your portfolio is mostly futile. When something comes out, that event is integrated into the price of relevant assets within seconds. Trying to parse through a constant bombardment of information just isn't necessary.

Political events… doesn't matter.

Price of oil… who cares.

Big economic data release… *snore*

The Fed… the who?

The only "active" part of the approach should be some type of periodic rebalancing to maintain the desired asset allocations. Overall, the risk in each portion of the portfolio should remain balanced. For example, if equities have been doing well and taking up more risk in your portfolio than they should, you can sell off a portion and invest the proceeds into assets that are underweight (or in the case of a contribution, add proportionally less to stocks than other assets). If equities have been doing relatively poorly, you would sell off other elements that may now have disproportionate weighting and put the money back in stocks.

The act of rebalancing provides an inbuilt "buy low/sell high" framework without actually making active judgments on what's good and what's not, as that's difficult to do. Some things are just impossible to know with any resolution. Plus you don't need to know what'll happen in the future to invest successfully anyway.

How often should you rebalance?

I used a sample (relatively risk-balanced) allocation of assets and backtested their performance from 1979 to the present. I used the following allocation:

  • U.S. stocks - 20%
  • High-Yield Corporate Bonds - 20%
  • Intermediate-Term Treasuries - 15%
  • Long-Term Treasuries - 15%
  • Long-Term Corporate Bonds - 8%
  • Long-Term Tax Exempt Bonds - 7%
  • Gold - 15%

I backtested using the following rebalancing time periods: monthly, quarterly, semi-annually, annually, and never.

The returns and graphs below simply assume $100,000 was invested in January 1979, with no further contribution or withdraw since, and with no leverage applied.

Monthly

By rebalancing monthly, over the course of 38-39 years, we had a compounded annual growth return ("CAGR") of 8.77% with a standard deviation in annual returns of 6.88%. Our worst year was -7.08% (2008) and maximum drawdown came to -15.55% (from November 2007 to March 2009).

This portfolio was profitable in 295 of 460 months (64.1%). It was profitable in 34 of the 39 years and only lost more than 3.1% in a year once (in 2008). (For anyone interested, the S&P 500 was profitable in 32 of the 39 years and lost double-digit annual percentage amounts four times - 2000-2002 and 2008).

Quarterly

If we had rebalanced quarterly, we can see that there was no functional difference relative to rebalancing monthly. Returns, volatility, worst year, drawdown, and Sharpe and Sortino (measures of risk-adjusted return) were all very much similar.

So right off the bat we can probably discern that rebalancing as frequently as every month might not be necessary.

Semi-Annually

Once again, very much on par with the monthly and quarterly rebalancing figures. Every statistic is relatively similar.

This suggests that rebalancing every three months for a strategy that is designed to be truly long-term - buy-and-hold over the course of decades, if need be - is probably too excessive.

Annually

And… no blurb needed. Annual rebalancing doesn't appear any worse than the monthly rebalancing.

Never

Now we actually do see a conspicuous difference. For a portfolio that was held for nearly 40 years, rebalancing any more frequently than annually showed no improvement in either returns or risk.

However, if assets are indeed held long enough, the relative weightings of each will become skewed, which will also in turn skew the risks in the portfolio unfavorably. The asset mix will no longer be relatively balanced. This means more reliance on a certain type of environment for the portfolio as a whole to do well.

For instance, if stocks do well over time and become a greater weighted proportion of the portfolio, the portfolio will become increasingly reliant on a higher growth, higher inflation environment (relative to embedded forward expectations). Given this portfolio is passive, and we aren't actively betting on what the economy will do in these aspects, we naturally want to balance frequently enough to the point where this won't become an issue.

The returns are higher. But the risk-adjusted returns, which is what we should ideally be after, suffer in a true "buy-and-forget" portfolio. Our annualized volatility increases by about 130 bps for only an additional 30 bps of annualized return. The Sharpe and Sortino ratios suffer accordingly.

Conclusion

For those investing passively with a long-term time horizon, there is probably no need to rebalance a portfolio several times per year.

I also ran this on various other portfolios that involve a mix of holdings across multiple asset classes. And it also made no real statistical difference to balance more than once annually. This also held true among shorter holding periods, such as the course of just one decade - i.e., rebalancing ten times (once annually) was just as effective as rebalancing more than ten times.

This is all, of course, a good thing. It requires less time, fewer commission fees, and can reduce tax bills. Investing doesn't have to be overly complicated - pick an allocation mix that is relatively balanced in order to reduce volatility and smooth out returns, balance about once per year, and one should do quite well over time.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.