Robeco portfolio manager Wilma de Groot explains the importance of research and discusses two major new projects in her department. “ESG inclusion can be seen as a form of risk reduction”, she says.
She manages two benchmark-related Emerging Markets strategies together with Tim Dröge and Michael Strating: Core Quant Emerging Markets, better known as Core, and Active Quant Emerging Markets (Active), which has a more concentrated portfolio. Their stock-selection model celebrated its 15th anniversary this year. It is fifteen years since Robeco first discovered that quantitative stock- selection techniques which were known to be effective in developed markets also worked in emerging markets.
We are currently working on two large research projects: First, we are looking at how to incorporate stock-specific trading-cost estimates further, to improve after-cost performance. These costs consist of estimated broker commissions, taxes, currency-conversion costs and market impact. In order to do this we have built a proprietary transaction-costs model which predicts the expected costs before we carry out a transaction.
The next step is to integrate the element of transaction costs into our investment process. This means that we take into account transaction costs when we select stocks. So a stock with high trading costs is less likely to be selected for the portfolio, or may carry a lower active weight. Another idea is to look at the trade-off between expected alpha and expected costs. A stock might have high trading costs, but may still be worth buying because its alpha is expected to be high.
Second, we want to incorporate ESG (environmental, social and governance) information further into our selection process for emerging market stocks. This is not about exclusion lists, but about ESG integration. This project combines the traditional strengths of Robeco: quantitative investing, sustainability and emerging markets and makes use of RobecoSAM’s ESG scores.
The idea here is to incorporate screening of ESG factors into our portfolio-construction process differently than we do now. The aim is to impose a minimum ESG threshold (e.g. always at least as good as the benchmark), while still maintaining performance. ESG inclusion can be seen as a form of risk reduction and complements our integrated risk-management philosophy.
We aim to remove those risks that are not rewarded with higher returns. The risk of being exposed to unsustainable companies is an example. Until recently it was very difficult to rely on ESG information for emerging markets because of the low data coverage. But this is steadily improving, and we now have data representing around 80% of the market capitalization from the MSCI Emerging Markets Index.
Our approach is different in three ways. First, we cooperate closely with the Robeco Emerging Markets Equities team that looks at the fundamentals in these markets. We apply their knowledge to improve data quality for the stock-selection model and to reduce trading costs. Take for example the Brazilian oil and gas company Petrobras; it has both a local and US listing for its regular and preferred share classes. We look at which listing and share class has the best data quality. This can depend on the number of analysts who cover it, for example. For trading purposes, we look for the most liquid share class in order to minimize trading costs.
Second, we adopt an integrated risk-management technique. We prefer to start incorporating risk management when defining our variables. For example, by avoiding traditional momentum strategies which become biased to high (low) beta stocks in rising (falling) markets and being highly vulnerable to market reversals. Integrated risk management dynamically reduces the most important undesirable time-varying style tilts and helps to stabilize the risk contribution of these variables over time. The common practice of completely postponing risk management until the portfolio- construction phase is less effective.
The third factor is the way we construct portfolios. We do not use statistical portfolio optimizers which build a portfolio based on the covariance between stocks. These optimizers might distort portfolio characteristics in unpredictable ways. Instead we use a proprietary portfolio construction algorithm that it is fully tailored to our needs. Our approach uses fixed active stock weights for selecting new stocks and avoids extreme weightings. Because our approach is different, we also avoid overcrowded trades.
‘We prefer to start incorporating risk management when defining our variables’
Core uses only the constituents of the MSCI Emerging Markets Index to determine its universe. This universe can change. For example, in June MSCI reclassified two frontier markets, Qatar and the United Arab Emirates (UAE), as emerging markets. These upgrades are not an issue for us, because our research has demonstrated that well-known proven factors in developed and emerging markets, such as value and momentum, can also be applied effectively in frontier markets.
Active can invest in stocks in frontier markets – countries outside the MSCI Emerging Markets Index – as long as they are constituents of either the FTSE Emerging Markets or the S&P/ IFC Emerging Markets indices. Active currently invests in Pakistani stocks. When frontier markets are upgraded to the MSCI EM index, we can leverage on our trading experiences there, and apply these to our Core strategy as well.