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.
Managing transaction costs is one of the major considerations in Robeco’s Quantitative Equity strategies, and over the years we have put considerable effort into finding ways to keep them as low as possible. In this article, Weili Zhou, a senior researcher in our Quantitative Equities team, describes the approaches we adopt to achieve this aim.
“Managing transaction costs involves dealing with a number of important issues. These include determining the optimal level of portfolio turnover, sizing individual positions and orders, executing trades in the most efficient manner possible, and making prudent forecasts of our strategies’ capacities.”
“One way to control trading costs is to limit turnover. We rank all the stocks in our investable universe, and of course we want to buy as many of the stocks in, say, the top quintile as possible. But suppose a stock drops out of the top quintile and is ranked at 21%. Should we immediately sell it, or keep it for a while? Based on our research and practical experience we have found that in general the alpha expectation for such stocks is still higher than for an average stock, so it doesn’t make sense to incur trading costs.”
‘One way to control trading costs is to limit turnover’
“We determine the optimal buy and sell thresholds for our quant strategies by taking into account the expected outperformance of the stocks we hold, the potential performance of others that are highly ranked, and the costs that these moves would involve. These thresholds are fully incorporated in our proprietary portfolio construction algorithm. As a result, we are able to keep the turnover of our quant strategies low by avoiding unnecessary transactions.”
“Let’s take Conservative Equities as an example. In this strategy, we hold stocks for an average of four years, so the expected annual turnover is around 25%, single-counted. This figure lies at the low end of the range compared with low-volatility funds managed by other companies, and in some years it’s even lower than the turnover of the MSCI MinVol index. This low turnover is important in helping to maximize the net returns that our clients ultimately receive.”
“Another important consideration in controlling transaction costs is mitigating both the fees involved in trading (fixed costs) and the market impact of our trades (implicit costs). To this end, we take into account both liquidity and trading cost estimates when we are determining the optimal position size for each holding. We measure liquidity as the average dollar-traded volume of a stock in the recent period, while trade cost estimates are calculated by our in-house trading cost model. Generally speaking, we are able to take larger (active) positions in liquid, cheaper-to-trade stocks than in illiquid, expensive ones.”
“On top of that, we have also developed our proprietary algorithm in such a way that it is able to slice and dice orders in order to minimize market impact while ensuring trades are completed within the intended timeframe. So instead of getting into a full position in one go, the algorithm may conclude that the best way is to build it up in steps at every round of rebalancing, as long as the ranking remains attractive.”
“Our proprietary trading cost model is something that sets us apart, as it enables us to estimate the costs of any trade and adjust our positions accordingly. The model takes into account two kinds of cost. First, explicit, fixed costs such as commissions, taxes, and currency conversions. Second, implicit trading costs, or in other words the market impact of our trades.”
“On the fixed cost side, the model incorporates our own fee structures, such as broker commissions. Our trading desk has been able to suppress brokerage costs a lot over the years. With program or algorithm trades in developed markets, our current execution-only commission is only around 1.5 basis points. This is less than what many of our competitors pay. In emerging markets we pay around 3 to 4 basis points for these kinds of trade, which is again at the low end of the spectrum.”
“In terms of trade impact, we built our model based on Robeco’s own trading history and thus our own unique trading style. We didn’t want to use an off-the-shelf model from a broker as these are calibrated based mainly on small program or algo trades – say between 0–20% of daily traded volume – and on beta trades such as index adjustments, which are very different to our alpha trades. In contrast, our trading database includes many large trades of above 50% of average daily volume, especially in emerging markets, in which we run around EUR 16 billion of assets. As a result, our model is well suited to make trading cost estimates ranging from 0–100% of average daily volume.”
“And our model works. There’s a good fit between its estimates and realized trading results for all kinds of stocks, from the largest to the smallest caps, from the most to the least liquid names, and for equities from both developed and emerging markets.”
“Our proprietary trading cost model is also fully integrated in our portfolio construction algorithm in a way that would be difficult to achieve with a third-party system. So when our managers are rebalancing their portfolios they can immediately see all the underlying commissions, taxes and market impact of their trades, as well as their total expected costs. This helps them to better understand the outcome of rebalancing (in terms of positions and order sizes), and to better monitor risks during execution.”
“We are fully aware that limits on position and trade sizes have an effect on outperformance potential. For example, if you identify a very attractive small cap that has poor liquidity, you might decide to invest less in the name or even skip it. This compromises your ability to generate alpha. Once a strategy grows beyond a certain size, the number of dilemmas like this that you face starts to increase.”
“This is why capacity management is so important. As a fund grows, it becomes increasingly difficult for it to deliver strong outperformance. On the one hand, trading costs go up, while on the other, alpha potential falls as the opportunity set shrinks. So a prudent asset manager should know where to draw the line in advance to ensure that their fund can deliver the investment results that it communicates are possible to its clients.”
“At Robeco, we’ve adopted conservative guidelines when estimating capacities, and communicate them clearly to our clients. In determining capacities, we not only simulate the impact of fund size on net performance, but also how much liquidity risk our portfolios would be exposed to as they increase in size – not just under normal conditions, but also in the kind of environment that investors all want to redeem at the same time. We have to make sure we could offer good returns as well as sufficient liquidity in such an environment. At present, all of our strategies have plenty of remaining capacity.”