Rebalancing versus buy and hold: How advisable is it to stick to the investment strategy?

Promising rebalancing strategy during the COVID-19 pandemic

Investors who have adhered to their investment strategies by rebalancing during the COVID-19 pandemic are probably seeing some gains. Consistent implementation tightly linked to one's investment strategy triggered the buying of riskier investments, such as equities, while selling off cash and bond positions. That means the speedy recovery on equity markets up until now has also generated additional returns for those sticking close to their investment strategies.

Simulation of a pension fund portfolio over the past 20 years

There are a variety of ways to execute a rebalancing approach. For the example below, we have chosen a monthly rebalancing option (periodic rebalancing) with the intention of allowing only minor deviation from the investment strategy in order to create a version of rebalancing that shows as much distinction from a buy-and-hold approach as possible. In practice, however, different rebalancing strategies are also used, and they allow wider drifts – that is, deviations – from the investment strategy (e.g. rebalancing when a trading range is broken). Therefore, they tend to more closely resemble a buy-and-hold approach in terms of how they work. The result is that fewer instances of rebalancing are required, and that lowers the transaction costs for portfolio reallocation.

Figure 1 shows the performance of two portfolios in the period from 2000 to 2020. The asset allocation with which they started resembled that of a pension fund: 30% Swiss equities, 30% CHF bonds, 30% Swiss real estate (listed), and 10% CHF cash and cash equivalents. The first portfolio pursued a buy-and-hold strategy. In other words, the performance of the asset classes affected its asset allocation. The other portfolio is rebalanced at the end of each month, always resetting it to its original (neutral) asset allocation (assumption: transaction costs of ten basis points). The right axis of the same chart also indicates the weighting of the asset classes with higher yields – equities and real estate – together on the respective dates. Figure 2 illustrates the performance of the asset classes viewed individually and based on the starting allocation.

Both approaches generated an identical annualised return of 4.0% over the period shown. The volatility of the rebalancing approach is higher, at 4.9% p.a., compared to 4.7% p.a. for the other approach. With rebalancing, the maximum potential loss is also higher, at -17.3% versus -15.6%. Based on the risk metrics, the buy-and-hold strategy would have required the investor to have lower risk ability compared to the rebalancing approach. It may also be astonishing that both approaches were extremely similar even though the equity and real estate allocation of the buy-and-hold approach drifted significantly from the neutral allocation multiple times. 

Possible conclusions from such analyses are difficult because a large number of parameters can have a significant impact on the results. Within the same simulation but with only two asset classes – 60% CHF bonds and 40% Swiss equities – as a starting allocation, the results change dramatically. In terms of yield, the rebalancing strategy now shows better performance (3.7% vs. 3.5% p.a.), is insteat it more volatile (5.5% vs. 5.0% p.a.). The rebalancing approach also has a higher maximum loss of -20.9% compared to -18.6%. With the buy-and-hold approach, the highest percentage of equities is now 47% and the lowest 24%.

Many deciding factors determine which approach is more successful

The reduction to two asset classes and the resulting change in performance reveal how sensitive the results are. Certain factors are relevant both when choosing what specific strategy to implement (buy and hold, rebalancing, or others) and when setting the parameters within that specific strategy (e.g. rebalancing frequency or size of any ranges). Those factors include, among others, the number of asset classes in the investment strategy, their weighting, volatility, correlation with the other positions in the portfolio, and developments on capital markets.

Another criterion that may play a role, particularly where strict rebalancing strategies are concerned, is transaction costs. In the above simulation, however, we see that the costs have a very minimal impact on performance. Not until the transactions reach the extremely high level of 50 basis points does the annualised yield of the rebalancing approach fall from 4.0% to 3.9%. During the current COVID-19 pandemic and especially where more illiquid asset classes are concerned, it has also been shown that the costs can soar to such levels and higher, at least temporarily.

Buy-and-hold has the advantage

Rebalancing is often recommended over buying and holding, particularly in crisis situations. The bearish periods on the equity markets shown in Figure 1 show that this strategy is not necessarily guaranteed to work better from a risk-reward standpoint. For example, the financial crisis proved that the buy-and-hold strategy, even at the beginning in late May 2007, entered the crisis with a higher proportion of equities and real estate than the rebalancing strategy (64% vs. 60%). By the time the crisis hit its low point in February 2009, buy-and-hold nevertheless achieved a better return (-15.4% vs. -17.0%).

An important reason for that is the decreasing equity allocation of the buy-and-hold approach over the course of the crisis. By the time the crisis had ended, the allocation was still just below 20% compared to the almost 28% using the rebalancing approach (returning to 30% before monthly rebalancing). The difference in returns between the two approaches consequently would be even greater if the simulation began at the onset of the crisis and thus with a neutral asset allocation. In that scenario, the buy-and-hold approach generates a yield of -13.8% compared to -17% for rebalancing by the time the crisis reaches its low point. Thanks to its higher equity holdings at the low point of the crisis, the rebalancing approach recovers its losses more quickly later, however. We also see a very similar progression after the dot-com bubble burst (2000 to 2002).

The third major crisis of this millennium is still ongoing. At the end of January of this year, the equity and real estate allocation of the buy-and-hold strategy was 71% (versus 60%), which led to a (slightly) heavier decline in returns by the end of March than the rebalancing strategy (-7.6% vs. -7.0%). The unfavorable overweighting of the buy-and-hold strategy in the riskier asset classes of equities and real estate at the start of the COVID-19 pandemic has had a somewhat limited effect because, among other things, the CHF bonds also posted a negative yield for the two months of February and March.

How advisable is the rebalancing strategy?

Judging by the risk-reward metrics obtained by the different approaches in the above crisis periods as well as over the entire time line, the buy-and-hold strategy generally performs better in this analysis. So, does that mean sticking to the investment strategy is not all that recommendable? It would be premature to reach such a conclusion or recommend giving preference to the buy-and-hold approach. How crucial the deciding factors specified above can be and how sensitive the results are can be illustrated once more with a look at the financial crisis. When it began in late May 2007, the proportion of equities using the buy-and-hold strategy was 32.6% (rebalancing strategy: 30%). If the previous market trend had resulted in a higher equity allocation of 36.5%, the buy-and-hold strategy would have suffered losses identical to those incurred through rebalancing. An equity allocation of, say, 40% in the buy-and-hold approach at the start of the financial crisis would have caused a decline that was two percentage points higher.

Investor benefit as an important argument

A second key aspect in assessing the two strategies concerns the benefit of the chosen approach to investors because, up to this point, the analysis has exclusively reflected an asset-based examination. Institutional investors, in particular, place great importance on investor benefit. Because they perform an essential task for society and therefore bear tremendous responsibility, they know their risk ability and yield targets precisely. That is the reason why they pay attention to keeping their investment strategy and investment portfolio (execution) in line with their investment profile (risk ability, yield target) at all times. If the portfolio risk exceeds the risk ability, it can have a far-reaching impact on the assets of their beneficiaries or insured.

With a stricter rebalancing approach, investors can better estimate in advance whether their investment strategy (which allows for fewer deviations compared to a buy-and-hold approach) will also enable them to survive difficult crises. Nonetheless, they might underestimate the extent of a potential crisis, but they will at least know with what asset allocation they have to overcome if they do. With a rebalancing strategy, it is also easier to determine the proper size of a buffer or reserves for such negative events than it is using a buy-and-hold strategy. This, in turn, reduces the risk of having to make adjustments in asset allocation in unfavorable moments (i.e. at the low point of the crisis). Pension funds, for example, try to absorb market fluctuations through their fluctuation reserves.

The better alignment of strategy and investor profile could ultimately be the main reason why institutional investors tend to choose a rebalancing approach. The analysis reveals that the additional risks (based on volatility and maximum losses) should be minimal in most cases using the buy-and-hold approach with a diversified portfolio and are even lower during major crises. Because of the wide variety of deciding factors, particularly the unknown capital market developments and the efforts of institutional investors to assess their (investment) risks accurately, it is, however, less likely that a strict buy-and-hold strategy will be used for the long term, and then only in extremely rare cases.