Portfolio rebalancing is sold as gospel to millions of investors by the financial industry. However, contrary to popular perception, rebalancing is a strategy with distinct advantages and disadvantages. Moreover, the way most investors rebalance their portfolios is not maximizing the gains that they could get from doing so and often serves to create unnecessary trading costs, taxes, and fees. If you have an investment portfolio of any size, knowing how to properly rebalance could retire you a few years early, or can improve your lifestyle in the interim.
For this article, I'm using a couple of papers to inform my thinking, one from Vanguard and one from TD Ameritrade. The TD Ameritrade paper, in particular, is quite informative. Note: if you don't follow everything in this article, go back and read these papers to help!
Understanding the logic of rebalancing
In order to understand rebalancing, there are some things you need to understand about the nature of investment portfolios.
1. Stocks have a higher return than bonds over time, and thus will grow to a greater and greater share of the portfolio as the years go by. As such, your portfolio will gradually drift towards the higher return and higher risk investments you've made. Eventually, a 60 stock/40 bond portfolio will become 90/10 (especially in today's low-interest rate environment). If this happens it's good because your equities have gone up, but it's bad because the standard deviation of your portfolio will rise over time, and you could lose 40-50 percent of your money right before you're supposed to retire. This also happens when you invest in successful technology stocks. Such stocks often go up so much faster than the rest of your portfolio, forcing you to contend with something called concentration risk, which you need to manage one way or another (let it ride a little but don't go all in). The classic solution is to rebalance your portfolio, which reduces risk, but in all likelihood reduces returns as well.
2. If you believe that asset class returns tend to revert to the mean over time, rebalancing can boost your total returns under the right circumstances. When one asset class outperforms, you can rotate into the underperforming asset class and boost your return. Whether this is true in practice or not depends on the time period you look at. The 2000s were a great decade to rebalance and you'd have outperformed buy-and-hold, but the 2010s haven't been as good. You'd have underperformed buy-and-hold during this period. If you believe, however, that markets trend over long periods of time, then you should know that rebalancing strategies do not necessarily reflect this view and will underperform in this sort of environment.
3. Market dislocations happen from time to time, causing stock prices to get temporarily out of whack. This is often caused by overleveraged traders getting margin calls and being forced to sell, but can also come from sellers unloading relatively illiquid assets and getting (temporarily) horrible prices for what they're selling, especially in small-cap stocks or junk bonds. If you've been investing for long enough, you've seen this happen over and over. However, most investors rebalance either quarterly or annually. This is a constraint and leaves the best opportunities to those who can follow the markets and take of opportunities any time they arise (or have free software that does it for them).
4. If you have assets that are negatively correlated with each other but that have positive expected returns, rebalancing is brilliant. An example of this is risk-parity US Treasuries (using either the 30-year Treasury ETF (TLT) or the better option (in a positive yield curve environment) of leveraged short term notes through the futures or Treasury repurchase market). US Treasuries tend to soar when everything else goes to hell in a handbag in a deflationary recession, meaning that there's more to investing a little in Treasuries than the 3-4 percent annual returns you can expect over the next 10-15 years. What happens when you own Treasuries is that you can use the big payoffs in bear markets to buy stocks when they've crashed and no one else is buying, and then make big money on the upswing. For example, TLT went up 34 percent in 2008. This gives you optionality, which you can use for big payoffs if you know what you're doing.
If you had used the money to buy stocks in 2008 from your TLT gains, you'd be sitting pretty. The upshot to this, however, is that if assets are strongly positively correlated, rebalancing isn't as likely to be as helpful.
1. The most obvious benefit of rebalancing is that it limits risk. If you set a portfolio allocation of 60/40, rebalancing the allocation with time will cushion equity risk. This is a near certainty.
2. As stated before, rebalancing can benefit from volatility, so if you get big market swings, you're usually buying low and selling high. As you can see from this graph, various rebalancing methods tend to be able to do exactly that. Note: The 0 percent band is to illustrate that the strategy is sensitive to transaction costs if overused.
Source: TD Ameritrade
3. Rebalancing gives investors a plan and may help them stick to their target allocations. This is a purely psychological benefit but a benefit nonetheless and one that may be overlooked.
4. When additional asset classes besides stocks and bonds are added to the mix, rebalancing becomes a better strategy. For example, domestic and international stock returns tend to revert to the mean over time once you account (or hedge) for currency fluctuations and control for the quality minus junk and value factors. If you're smart enough, you can do interesting stuff by playing currencies against stocks and bonds of their respective countries. For example, you can hedge out the euro from European stock investments and earn the difference in interest rates. So, if you're American and buy $100,000 worth of European stocks to diversify, you can hedge the Euro exposure and simultaneously boost your return and reduce your risk (you'd pocket the 2.4 percent difference between the swap rates and would net out the long euro position from the European stocks with a short euro position in the futures market).
1. It's not rocket science to understand that bond returns are going to be lower over the next 20 years than they were over the last 20. Here's a graph of 10-year Treasury rates over the last 20 years.
Source: St. Louis Fed
While bonds still provide a countercyclical element to your portfolio, they used to return 6+ percent, and now they return >3 percent if yields remain the same. As such, the higher your bond allocation, the lower your return is likely to be. By contrast, equity valuations aren't low, but they aren't historically high either, despite what the media seems to think. US equities are likely to return 9-10 percent going forward in USD terms, based on the current earnings yield and conservative nominal GDP and per share profit growth assumptions. Also notable is the fact that bonds are incredibly margin-friendly, meaning you can typically get bond exposure equal to 100 percent of your portfolio for 2-5 percent initial margin in the futures market. This could free up cash for both a higher allocation to equities and a larger cash cushion in your portfolio.
2. Rebalancing creates trading costs and fees. While the risk-reduction benefits of rebalancing are a certain benefit, trading costs are a certain loss, as are taxes, if you are selling stocks or bonds for a capital gain. Most investors would probably be curious to know if the benefit outweighs the cost. The answer is it depends on how you rebalance.
3. Rebalancing may paradoxically make it harder for you to achieve your investing goals (if done improperly) by reducing your compounded annual rate of return. If you're 25 years old and single, you shouldn't be worried about volatility the way a 50-year-old self-employed business owner looking to retire in a few years is.
4. You probably shouldn't be rebalancing your individual stock positions the way you'd rebalance asset classes. A lot more stocks than you would expect do go to zero, so applying asset class rebalancing principles to individual stocks can wreck your portfolio.
How to rebalance the right way
Source: TD Ameritrade
After knowing the pros and cons of rebalancing in general, should you rebalance? I believe that the answer is yes, but that the way that most investors are going about rebalancing is wrong.
For one, you have the momentum effect in stocks, which makes sense when you consider that a ton of investors just rebalance quarterly and sell what's been working, even if the economic fundamentals support it. I write a lot about exploiting the constraints and biases of the investing public, and this is yet another way to do this. If investors rebalance too often or the wrong way, they're going to sell winners and buy losers, which has been shown to cost money.
At the same time, you have mean reversion, which often arises from stocks being sold hard for no apparent reason. You see this with frequent flash crashes in US equities, the most famous being 1987 crash, where equities dropped over 20 percent in one day. While there are better regulatory and exchange safeguards against markets randomly crashing today, limiting yourself to rebalancing quarterly is reactive rather than proactive. Trading profits will accrue to those who are willing to take opportunities when the market gives them up. Interest rates also are known to moderately revert to the mean over both long term and short term periods.
So, how do you optimize rebalancing strategies?
Here are a few tips that I can think of that can help you achieve your goals.
1. The first way that I can think of is to systematically invest your stock dividends into bonds. If bond returns are 4 percent and stocks are 9, then taking maybe 2 percent per year in dividends from the stocks and putting them in bonds makes the allocation gradient from bonds to stocks disappear. Now the position size growth rate of your stock position is 7 percent, and the position size growth rate of your bond position is 6 percent on average. This has the advantage of dramatically reducing your risk and transactions cost while still leaving opportunities open when market dislocations happen in stocks to buy stocks for rock bottom prices. Yes, this likely reduces your returns but also gives you a better Sharpe ratio. If you have access to inexpensive leverage, you use this setup to get higher returns for the same level of risk.
2. The TD Ameritrade paper found that the optimal rebalancing strategy was a 20 percent rebalance band (with a 10 percent tolerance band). They also used 5 asset classes for their portfolio. I'll illustrate how this works. If you have 20 percent of your portfolio in US stocks, you would rebalance if it hit 24 percent (24 minus 20 is 4, and 4 is 20 percent of 20). You wouldn't go back to 20 percent, however. You'd only rebalance to 22 percent. Why do this? For one, it reduces transaction cost, instead of having to move money in every asset class every time you rebalance, you let it ride a little to your tolerance band. Usually, you'd only have to make 2 transactions in this setup, rather than one per asset class, which would be 5 transactions. Additionally, research shows that winners tend to keep winning, so this gives you a passive exposure to the momentum factor. If an asset class insists on outperforming the others, it will get a higher weight as long as the risk isn't out of whack.
3. Find assets that are negatively correlated with each other. Total bond return funds like (BND) are correlated positively with equities due to their corporate bond exposure. Treasuries and gold tend to be negatively correlated in times of stress. Risk parity helps maximize the gains from rebalancing by magnifying negative correlation/positive expected return combinations of assets.
4. Intuitively, using rebalance bands rather than rebalancing on a schedule makes you less susceptible to everyone else doing the same thing you're doing. This allows you to profit from market dislocations rather than play into them. I like the TD Ameritrade rebalancing strategy and will likely use it in modified form in my investment model.
5. Rebalancing gains from TLT outweigh tax costs for those below the top marginal tax bracket vs. municipal bonds (MUB) due to risk parity and negative correlation with equities. For those in the top tax bracket, it's a little more complicated, but the cost can be partially managed by shifting some Treasuries into tax-deferred accounts and keeping municipal bonds in table accounts (no need to shift all, however). Also, you can work around some of this by using the futures market for favorable tax provisions as well as the carryback loss provision and ability to netting against all other futures gains/losses on one P/L. As always, talk to your CPA!
Done right, rebalancing boost the annual average returns of a typical investor by 0.5 percent or more per year. Experienced traders have greater opportunities to take advantage of market situations. Done wrong, it will cost you money in taxes, fees, and trading costs and the costs will exceed the gains. In particular, knowing how to properly rebalance has helped inform my views on my macro trading model, and should be able to boost the returns in my macro risk-parity model by 1-1.5 percent per year.
Did you enjoy this article? Consider following me for future research updates!
Disclosure: I am/we are long TLT, SPY, VYM. 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.