Disclaimer: I'm learning how to invest. I'm sharing the findings I've come across so far in my study of this particular problem.
Remember that risk and return are two sides of the same coin. There are no risk-free investments. Your portfolio's composition is one "control knob" (maybe the only one) you have to dial up or down the risk of your investments.
Rebalancing is probably more about re-evaluating your risk tolerance (and diversifying assets) than it is about maximizing your returns. If it was only about purely maximizing returns in the long-run, then rebalancing would probably boil down to having 100% stock all the time.
I'm assuming that your portfolio is a mixture of equity and fixed-income assets (e.g. bonds). Each of these assets grows and compounds at a different rate. Equity (stock) will tend to grow much larger over time than bonds or GICs, for instance. Just to illustrate, from the historical period between 1900 to 2021, in the US, equities (stock) have shown an annualized return of 9.7% per year versus 5.0% on bonds, 3.7% on bills, and inflation of 2.9% per year.
[Also, ] Over the 121 years [from 1900 to 2021], an initial investment of USD 1, with dividends reinvested, would have grown in purchasing power by 2,291 times. The corresponding multiples for bonds and bills are 12.5 and 2.6 times the initial investment, respectively.
Source: Credit Suisse Investment Returns Yearbook 2021 -- chapter 2
So, naturally, if on year 1 you made an initial lump sum investment into two separate accounts, 60% in a stock account, and 40% into bonds, you're bound to see their relative value spread apart over time.
About that pesky "risk" thing I mentioned earlier... yes, stocks will tend to have outstanding returns over time, but they can also have devastating returns in the short term (and they've shown that, several times over, 2000-2002, 2008, 2009, 2022, etc.). Companies come and go, bubbles burst, markets crash, and so on. So you have to evaluate your situation and weigh in the risk factor. Ask yourself, "what would happen if my stocks went down 50% tomorrow and it took 5 years to come out even with my initial investment?". Bonds have had bad years too, but nowhere as much as stocks.
After year 1 my allocation (because of incredible equity performance) becomes 80-20... Possibly my initial asset allocation was not correct and I really want an 80-20 split.
The fact that your 60-40 portfolio mutated into 80-20 does not reflect a bad choice on your part. If you'd initially picked 80-20, it might have become 90-10 a year later. It could be that one side stayed the same, and the other one increased or decreased, or that they both went their separate way. You've got two unpredictable race dogs running around, but their leashes have different lengths.
Rebalancing gives you an opportunity to re-evaluate your risk tolerance
The ratio you should have in your aim (20/80, 40/60, 80/20, or 100/0, etc), is one that provides the returns what will let you achieve your goals, while staying under a risk level that you can tolerate. Your goals will change (e.g. maybe you need that money in 5years time for a downpayment), so it's not like you will pick that "magic" ratio once, and stick with it forever. If you are approaching retirement you might want safer investments. If you landed on a juicy inheritance, you might feel like you can handle more risk. If you changed to a job that provides a good pension, maybe you can start affording more risk. You might decide, for instance, to follow a strategy where your bonds percentage matches your age in years (or your age minus 10 or 15 or even 20, if you can stomach it).
That said, if your goals haven't changed form year 1 to year 2, then rebalancing is the way you have to "stick to your guns" while keeping the statistics on your side.
Rebalancing helps you keep your demons in check
A potential side effect of rebalancing, a somewhat psychological one, is that it will make your portfolio look more predictable over time. Humans tend to make more mistakes when they get surprised. And so, less volatility will help you avoid irrational moves. It will help you combat the (irrational) impulse to add more money into a stock that is climbing, or conversely, urgently selling a seemingly-sinking one.
A "normal" stock year is filled with ups and downs. We tend to be optimistic. When we lose, we think it’s bad luck, and when we win, we instead tend to think it’s because of skillful decisions (luck? no way. i called it!)
Sure, you can predict how you'll feel if you lose a large percentage of your money, but it's hard to know for sure how it's going to really feel until it actually happens.
Rebalancing, during those events, actually is a good strategy to fight your demons and will generally help you (indirectly) "buy low, and sell high". It's a strategy that has been proven to be beneficial over time, over say the "cowboy approach" to investing that relies on market predictions.
Rebalancing is roughly equivalent to selling all your assets in year 2 and reallocating them as if it was now year 1. If you thought 60/40 was your ideal risk tolerance, it probably is a good idea to pick the same point the next year.
(But of course, during rebalancing, you’re not selling everything and rebuying. You'll want to also minimize the amount of actual movement your money has to go through to reach your desired target, to save on taxes and fees).
Caveat 1 to rebalancing: diversification
In order for any rebalancing to be effective, you assets must be diversified, meaning they should grow or shrink more or less independently (or be inversely correlated). Two such assets that have shown to work well together historically are quality-bonds and equities.
To illustrate why it matters, suppose you were to hypothetically put 50% of your money into a (fictitious) fund of all stocks beginning with the letter "A", and 50% of your money in a (fictitious) fund of stocks beginning with the letter "B", it's not clear that rebalancing would be effective in reducing your risk. You'd have to make sure that the companies forming the two funds are relatively complementary, across different regions, and different industry sectors.
If you had 60/40 in this fictitious fund mix, and it became 80/20 a year later, that in itself might not be a valid reason to justify rebalancing at that moment. You'd have to monitor precisely how the fate of the two funds are intertwined on year 2, to assess the risk properly. (Maybe, for instance, new companies named “A”pple and “A”mazon start dominating one half of your portfolio, and whenever they go up, your bitcoin fund “B”TC on the other side goes up with it (or crash together)).
See "Modern Portfolio Theory".
Rebalancing caveat 2: fees
Consider that rebalancing has a cost in transaction and management fees (and a time cost in your schedule too). So, there is some strategy to it regarding “when”, or “how” to apply it. There's typically a cost to placing transactions (in the form of bid-ask spread or ETFs brokerage commissions), and so if you keep micro-adjusting every picosecond, you’ll do nothing but rake up fees that totally cancel out any gain you might make.
Some investors elect to just rebalance their assets whenever they inject new money (or retirees when they withdraw) .
If they have a hypothetical fresh $1000 to invest, they'll place a greater portion of it on their fund that has lagged behind, but leave the previously-invested money as is.
Depending on taxability of the account, you might get a tax hit if you sell. So, sometimes it's better if you can rebalance by just buying versus a combination of buying and selling. (Or conversely, if you're retired and sell more than you buy, you might be able to rebalance by selling only from one side of your portfolio). That approach is particularly appealing if your online brokerage discounts fees on purchases.
This works well when you have a small sum invested, but if you add $1000 into a million dollar pool, you might find that the relatively small injection of new cash won't affect your current investment allocation ratio very much. There's another famous strategy called the “5/25 rule,” in this case can provide a rule of thumb use to determine when it's worth rebalancing.
I haven't come across studies (yet) that settle the score on the ideal frequency for rebalancing. There's no solid evidence that says that a particular fixed interval is definitely better. But on years with more fluctuation/wide swings, it helps to review and rebalance more often. Many passive allocation indexed funds will help make some of those decisions for you (more on this in my next point)
Rebalancing caveat 3: it's complicated
Sometimes rebalancing can be quite complex, especially when your asset classes span different accounts with different tax implications, and/or your investments have different maturities. E.g. maybe you have your stock in a tax free account, but your contributions in that account are limited to a certain amount per year. Or maybe your fixed income assets have terms on them (e.g. GICs), and they're not as liquid as you'd like them to be.
For this reason, some investors choose to invest in fixed-mix allocation index funds which do the rebalancing for you and will keep your desired ratio (80/20, 60/40, etc). They save investors the work of manually rebalancing. That can take some of the complexity away, for a minimal management fee (e.g. ~0.10% for Vanguard's funds) over what you'd pay if you bought the different classes independently (you might need to rebalance between a patchwork of 5-6 different funds to replicate the same portfolio composition).
- A lot of my theory here is inspired by this book: Bortolotti, Dan. Reboot Your Portfolio: 9 Steps to Successful Investing with ETFs. Milner & Associates Inc.
Relevant questions on this site:
Rebalancing vs Reacting