Robeco Institutional Asset Management B.V. (Dubai office) is regulated by the Dubai Financial Services Authority (“DFSA”) and only deals with Professional Clients and does not deal with Retail Clients as defined by the DFSA.
Neither information nor any opinion expressed on the website constitutes a solicitation, an offer or a recommendation to buy, sell or dispose of any investment, to engage in any other transaction or to provide any investment advice or service. An investment in a Robeco product should only be made after reading the related legal documents such as management regulations, prospectuses, annual and semi-annual reports, which can be all be obtained free of charge at this website and at the Robeco offices in each country where Robeco has a presence.
Factor-based allocation has become increasingly popular in recent years. But how to implement it in practice still remains a puzzle for many newcomers. One concern often voiced by investors is avoiding excessive portfolio rotation.
One oft-heard criticism of explicit allocation to factors is that it inevitably leads to high, or even excessive, portfolio turnover. Indeed, while following a cap-weighted market index can essentially be seen as a ‘buy-and-hold’ approach, with limited portfolio activity, explicit allocation to factors necessarily leads to more dynamic trading.
In cap-weighted indices, stock weights fluctuate naturally with the prices of constituent securities, and changes in the portfolio composition are only triggered by large changes in free-float capitalizations or corporate actions such as mergers, splits or new listings or delistings.
On the contrary, factor-based investment strategies generate turnover from the periodic rebalancing required to maintain optimal exposure to the targeted premiums, for example value, momentum, low volatility or quality. This necessary turnover has led many academics and investors to question whether factor-based solutions are really worthwhile, given the higher trading costs associated with these strategies.
For example, a recent study(1) by Research Affiliates warned about the significant slippage between the factor returns realized by mutual fund managers and the theoretical factor returns that would have been achieved by virtual portfolios, over the 1991–2016 period. The authors attributed this gap to a number of costs related to implementation, including trading costs.
In a 2016 white paper(2), Joop Huij and Georgi Kyosev, of Robeco’s Quant research team, warned specifically about the high rebalancing costs implied by the replication of some common smart beta indices. Analyzing the impact of composition changes for two popular indices, they found that these costs are actually higher than they might appear, as the rebalancing process also leads to lower index returns. This is because strategies that follow publicly available indices, for which changes are announced in advance, tend to buy stocks that have just had a price run-up, and sell stocks that have just suffered a price decrease.
More generally, an FTSE Russell survey carried out in 2016 suggested that avoiding excessive portfolio turnover ranked fourth among investor concerns, when considering factor-oriented allocation.
But while the risk of excessive turnover should not be overlooked, it should not be exaggerated, either. In fact, it is possible to considerably reduce turnover without hampering performance too much. Robeco’s in-house research shows that when investors start keeping securities with less attractive factor qualities in their portfolios for longer, trading costs tend to decrease faster than the gross return. As a result, the net return/risk ratio tends to increase when turnover starts to decline.
‘Turnover can be reduced without lowering gross returns too much, but only up to a certain point’
This finding does not mean that portfolio changes should be minimized. Turnover can be reduced without lowering gross returns too much, but only up to a certain point. And gross returns also tend to drop rapidly once we allow unattractive securities to remain in the portfolio for too long or rebalance too infrequently. Investors must therefore find the optimal trade-off between factor exposure and rebalancing costs, in order maximize after-cost performance.
There are many ways to reduce and control portfolio turnover, and that can be applied to all kinds of factor-based strategies. The most obvious one is setting and adjusting fixed rebalancing intervals, in order to reassess factor exposures more or less frequently. Another option is allowing a portfolio to deviate more or less from its ideal composition, if only factor exposures were taken into account and implementation costs were neglected. The greater the deviation tolerance, the lower the turnover will tend to be.
In addition to these general techniques, which are widely used by investment managers and index providers, there are also more strategy-specific ways to reduce turnover. Empirical studies carried out on the short-term reversal phenomenon, which has been extensively documented in the academic literature, provide a good illustration of this.
Short-term reversal strategies exploit the fact that stocks that experience huge gains or losses during one month tend to reverse that trend the following month. However, many investors remain skeptical about this kind of approach because they involve huge turnover, as signals typically change completely every month.
But a 2011 paper(3) by Wilma de Groot, Joop Huij and Weili Zhou, of Robeco’s Quant Equity research team, showed that the high transaction costs incurred in many these investment strategies implemented in the US stock market could largely be attributed to excessive trading in small caps. Trading costs could therefore be significantly reduced by limiting the stock universe to large caps. Similarly, comparable ways to reduce turnover can often be found for different kinds of quantitative strategies.
All of Robeco’s quantitative strategies use portfolio-construction processes designed to keep trading low and trading costs under control, using a securities-ranking approach. This kind of method is less sensitive to changing market inputs. Moreover, for credit markets, which lack the immediacy seen in equity markets and where keeping transaction costs under control proves more challenging(4), we have developed a specific investment process, in which liquidity management is actually embedded in the portfolio construction process itself. This enables us to send only those orders which have a high probability of being executed.