The general theme of last week’s missive was that good investing comes from a starting point of trying not to be wrong, rather than trying to be 100% correct constantly. Rather than intellectually trying to hit sixes out the stadium and eventually getting caught out, that we should be comfortable to nudge a few singles and keep the scoreboard ticking over.
One of the main ways that we can keep our investments in line with our needs and goals is to “rebalance it”. Rebalancing our investments simply means placing trades to bring our current asset allocation back in line with our starting one.
Let’s say that we start off with a fairly common asset allocation of 40% in bonds and 60% in stocks1. Over time, stock markets are either going to outperform bond markets, or the opposite will be true and bonds will do better than stocks (equities).
In either case, our starting allocations to bonds and equities will change and start to drift away from what they originally were.
The maximum temporary drawdown (worst case scenario) historically for an 80% stock & 20% bond portfolio is around 40%, while it has been approximately 30% for a 60% stock & 40% bond portfolio. So we can see how, without some upkeep and attention, our portfolio may become an altogether different animal over time.
This being the case, and assuming that you chose your starting asset allocation for a reason, the question becomes “when and how should I rebalance my portfolio to bring it back into line with the original strategy?”
With the help of the folks at FE Analytics I have put together three different portfolios - all with a starting asset allocation of 60% in global equities and 40% in US Treasuries (US government bonds):
The portfolio is rebalanced back to its starting asset allocation quarterly (in March, June, September and December);
The portfolio is rebalanced annually in December each year; and
The portfolio is allowed to run from its start date in 1980 without rebalance.
Source: FE Analytics. No investment or advice fees are deducted from these returns.
First thing to say - there isn’t a lot of difference. All three do amazingly well.
The worst performer is portfolio number 3 (“no rebalance”) - which in truth surprised me a bit. A bit of a performance gap opens up around the Great Financial Crisis, which suggests that the other two portfolios’ ability to rebalance during this incredibly volatile period (systematically selling bonds which had gone up, to buy stocks at depressed prices) added real value.
Anytime that we look at performance over such long time periods though, the returns are so good that it can be difficult to “dig in” and see where the discrepancies are. The numbers on the y-axis get a bit silly.
So, I also looked at how each strategy performed over rolling five years timeframes, again starting in 1980. The below table shows the percentage of times that each strategy was the best performer over these rolling five year periods.
Source: FE Analytics. Again, no fees or charges are deducted from investment returns. Rolling five year, month end returns were used.
Over almost half of the five year performance periods, taking a strategy of never rebalancing was the clear winner. This is kind of more what I expected to see. Equity markets rise over time so if you let your portfolio run, the equity content is likely to get higher and returns get better.
It is also probably no surprise that the “never rebalance” portfolio tended to shine during periods of strong equity market returns, while the other two portfolios did better in the aftermath of some pretty rubbish periods for stock markets (post GFC and post “dot com bubble” most notably).
But there is a cost to adopting the no maintenance approach. If we look at the numbers one final way, and incorporate how volatile each strategy was then we can get a better insight into what the journey “felt like” for each.
Source: FE Analytics. Again, no fees or charges are deducted from investment returns.
The “no rebalance” portfolio exhibits the highest levels of volatility i.e the value of this portfolio moves around more than the others, although it isn’t a huge difference.
Again, this seems logical as equities tend to be more volatile than bonds, and a rising equity content within the “no rebalance” portfolio would mean higher volatility over time.
Over the full time period that we are looking at at least, both rebalancing strategies seem to offer somewhat of a “free lunch”. The rebalanced portfolios have generated better returns, and offered a smoother journey. The main downside being that they have also gone through long periods where they have done worse than the no maintenance option.
I suppose that this is to be expected, there should always be some cost to buying insurance. And rebalancing is a form of behavioural insurance - once I have determined what the right allocation is for me, I will every month/quarter/year come hell or high water rebalance back to that starting allocation.
Because the rules are established in advance, when a period of high emotion comes I know what to do. It wouldn’t have felt all that clever selling bonds to buy stocks in March 2009, but as we have seen from the above it was absolutely the right thing to do.
Whether you decide to rebalance your portfolio quarterly or annually doesn’t seem to make much difference. There are no trading costs incorporated into the above numbers, and so in the interests of keeping such costs low I would suggest that annual rebalancing is perfectly line.
But we are talking about semantics here. The important thing to do is to make a decision, have a plan and stick to it. I promise that that will set you apart from 99% of investors out there.
Next week, on the theme of regret minimisation, I’m going to set out what to do when you have one investment which makes up an excessively large proportion of your overall portfolio.
As ever, past performance is not indicative of future returns and none of the above is intended to constitute advice to any individual.
Comments