APG states that combining separate quantitative and fundamental strategies has clear benefits.
The Dutch asset manager APG is one of the largest investors of pension fund money in the world; it has over EUR 359 billion in assets under management and is responsible for 30% of all Dutch collective pensions. Quantitative equity strategies are especially important to APG.
We talked to Ronald van Dijk, APG’s head of developed market equities. He is responsible for investing EUR 65 billion in developed market equities on a quantitative basis, and EUR 40 billion on a fundamental basis. As an experienced investor he provides insight into how quantitative and fundamental strategies fit together. He also explains what institutional investors should keep an eye out for when using quantitative strategies.
We believe that there are clear diversification benefits in combining both separate strategies.
Our quantitative investments outperform at different times than our fundamental strategies. This enables us to have a more stable alpha than if we were to focus solely on one type of strategy. The factors that boost market returns vary - sometimes the market is factor-driven or sentiment driven which makes it particularly suitable for quantitative strategies. On the other hand, markets where dissecting hard to quantify elements such as changes in business strategy and external risk is important are more suitable for fundamental strategies.
This diversification exists because the strategies look at different aspects of a company and market sentiment. The investment horizon is also different - for quant a maximum of three years, while most of our fundamental strategies focus on three to five years. But some have a ten year horizon or even an indefinite horizon. Furthermore, our quantitative strategies often follow value strategies, while our external fundamental managers tend to follow a growth strategy. Sometimes value is in vogue, while other times it is growth.
Because of the size of our fund our only major restriction is deciding on what works. We do not have to limit ourselves to only a few strategies. We can add an extra strategy as long as it contributes to a more stable outperformance and money alpha.
Our portfolio management process gets better because fundamental and quantitative strategies are combined on one floor. There are not many asset managers else around that can offer this combination, but Robeco is one of them.
The techniques that were developed for quantitative strategies such as multi-faceted risk management, transaction costs minimization, and constrained portfolio construction can also be used for our fundamental equity strategies.
The term ‘drawbacks’ is too strong, but quantitative strategies are different from fundamental strategies. You need to be aware of the possible shortcomings. The depth of company analysis is less than that achieved with fundamental research. But this can be overcome by building a more diversified portfolio.
And the drawbacks of quant are diminishing. In the past the strategies only worked with structured data. For example, company data derived from balance sheets and analyst forecasts. Nowadays, the strategies can use texts as input for analysis. Databases with hard data are no longer a requirement. These texts can be analyzed to assess the current sentiment with regard to a particular stock or to evaluate the latest ESG risks.
The use of unstructured data is only in the initial phase. There are significant opportunities because of the increase in computing power and the use of econometric models. We cooperate with universities to assess all the options available.
Another drawback of quant investing that is disappearing is portfolio turnover. In the past a quant strategy was thought to only be possible in conjunction with high turnover. This is no longer the case. Our portfolio turnover is lower than many fundamental strategies. The quant signals are also valid for a longer period of time and our portfolio construction process aims to further reduce turnover.
Transparency, lower costs and a better risk-reward profile than market-weighted strategies. There are three different types of strategies we use:
It is an academic debate. Are they a compensation for risk or are they an anomaly of the CAPM theory? But either way we are happy. Even if they are compensations for risk, it is not a problem as long as we can get higher returns. We can diversify risk by combining different quant and fundamental strategies. Here it is important that the fundamental strategies do not have the same exposures to factor premiums as the quantitative strategies.
We are alert and continuously monitor how well signals work and whether or not a particular strategy is becoming overcrowded. Take for example, low-volatility investing. We closely watch how much is invested in these strategies. We are not concerned now, because it is, surprisingly, still a small percentage of the active investment strategies in the world. But if we feel a particular strategy is becoming overcrowded, we can reduce our exposure.
Dutch pension funds, including most of our clients, are quite happy to invest in low-volatility strategies; we have invested EUR 12 billion in them to date. This high degree of acceptance is also a result of the extent of academic research that has been carried out in this area. Dutch researchers made significant contributions to this academic evidence.
It changed pension funds’ perception of risk. For Dutch pension funds and also internationally, tracking error was very important. But this situation has changed. Total risk is now more important. How stable is the funding ratio: how much do the assets fluctuate in relation to the liabilities? That’s where the minimum volatility strategy comes into play. The offer of a more stable funding ratio is obviously very appealing to pension funds.
The concept is good. It gives an indication of how particular smart-beta indices perform. But there are also drawbacks. For us it is difficult to follow this benchmark closely, because the capacity is too limited for us. Therefore, we needed to find alternatives. Moreover, we want to combine minimum volatility with valuation aspects. Another reason is that we want to include ESG factors, which aren’t included in this index.
Our pension fund clients expect us to integrate ESG into every asset class. We have put a lot of research effort into how one could integrate ESG within quant. The quality and market coverage of ESG data is continuously improving. This creates great potential for combining ESG and quant in innovative ways. First and foremost our approach has to be aligned with our and our clients’ Responsible Investment Policy: which ESG aspects are important to us as an investor? Quants are experts in finding out how ESG can be used to contribute to risk and return.
Absolutely. The Dutch asset management industry has a several decades-long tradition in quant equity investing. It is a pioneer in quantitative techniques such as multi-factor investing, integrated transaction cost control and asset liability management. We are pleased with the number of good econometricians coming out of Dutch universities - and also in other relevant disciplines such as mathematics, physics and computer science. You need good people to successfully pursue quantitative strategies. Furthermore, the Netherlands can attract talent from abroad, because it has such a good reputation in quant.
A middle-of-the-road approach is not possible for quant investing; you either do it well or not at all. It requires long-term investments in people and a dedicated IT infrastructure. We hope the Dutch asset management industry will remain strong as this will help the exchange of new ideas and to attract talent to the industry.