Pension funds face two major challenges. They need to increase returns. The majority of pension funds have a coverage ratio which is only just above the minimum requirements. However, at the same time they have less appetite for risk, because they cannot afford to let this ratio fall further. Is factor investing the solution to meeting these seemingly opposite challenges? Head of Factor Investing Joop Huij specializes in empirical asset pricing and equity strategies and answers ten questions from pension funds.
(1) What are the benefits of using factor investing for institutional investors?
“The benefits are reducing risk and enhancing returns in the portfolio over longer periods. Furthermore, factor investing helps asset owners obtain greater insight into the under¬lying factors relating to risk and return. This enables them to construct robust portfolios designed to withstand shocks. Institutional investors also appreciate factor investing’s academic basis. The demand for hard evidence that an investment style actually works has increased as a result of the crisis. Investing without clear rational reasons is a thing of the past.”
(2) When does a factor really work?
“Many studies on new factors are based on past returns and make a lot of assumptions. For example, that there are no trading costs, that you can trade just as easily in small caps as in large caps and that there are no taxes. In practice, many of these new factors simply do not work. Therefore institutional investors should be skeptical when considering incorporating new factors in their investment strategy.”
“The factors that do work have two things in common: there is ample empirical evidence supporting their existence and there is an economic rationale for it. I was one of the authors of a study that looked into the performance of mutual funds that were exposed to factors. We found that the most well-known factors, value, momentum and low volatility worked best, while the more exotic factors did not.”
Many are based on past returns and make a lot of assumptions.
(3) What about the quality and the size factors?
“Quality has become a very popular factor. However, it is often not completely clear what quality is. The definitions used by investors and researchers vary and often include many different variables. For a variable such as return on equity there is very little academic evidence. What we see is that this variable is sometimes augmented by low volatility or value to increase performance. In reality it is then just another factor in a different package.”
“However, having said that, I do not dismiss quality as a factor entirely. There is empirical evidence for some new quality variables. But more evidence is still required for it to be fully acknowledged as a factor.”
“We do not consider size or small caps as distinct factors, although there is some empirical evidence for these. At the same time, we do advise institutional investors to work with a universe that also includes smaller-caps when implementing factor investing strategies for equities.”
(4) Do you expect premiums to disappear if more investors embrace factor investing?
“In theory, if more investors were to embrace the concept, it would push up asset prices in the most attractive segment of the market. So future returns would go down and the premiums would disappear.”
“But I do not expect this to happen. While large institutional investors are embracing it and allocating large sums of money to factors, many other investors are unable to engage in factor investing strategies because of organizational barriers. If you wish to adopt factor investing you would typically have to make substantial changes in your organizational setup, responsibilities and in the alignment of interests.”
(5) What is the difference between factor investing and risk premium investing?
“Investing in risk premiums and smart beta investing are frequently used terms that both refer to the same idea as factor investing: strategically allocating to factors. I prefer to avoid these terms, because both specifically indicate that factors are a compensation for taking risk. However, there is very little empirical evidence to suggest that you actually need to take this risk to generate returns.”
“There are alternative interpretations other than risk to explain the existence of these premiums. Take for example the behavioral biases of investors that can distort asset prices. Robeco believes that factor premiums are related to rational behavior such as the career concerns of portfolio managers. Their career prospects might be jeopardized if they significantly underperform for a considerable period of time after deviating too much from the benchmark.”
“For us it is very important to know more about the reasons why factor premiums exist. A more in-depth understanding can enable you to gain a better exposure to them.”
(6) What is the best approach if you wish to capture factor premiums?
“The choice for factor investing is the right one, but the degree of success depends on proper implementation. Factor investing works, even with a generic or index approach. But it is possible to capture factor premiums more efficiently by setting up a high conviction approach. If you implement factor strategies well, you can earn higher returns, reduce risk and cut trading costs. A good factor strategy avoids three things: unrewarded risk, going against other factor premiums and unnecessary turnover.”
It is very important to know why factor premiums exist.
(7) Should you time factors by changing the weights to the individual factors?
“First and foremost you can add value by diversifying across factors. Moreover, we found that it is very difficult to time them. The academic literature on the subject is not conclusive, although there are some studies showing the role of the business cycle on factor returns, and the impact of factor momentum. There are two important obstacles in timing factors.”
“First, timing factors typically leads to higher transactions costs, which can have a large impact on total returns. Second, there are only a limited number of investment choices to be made. This phenomenon is called limited market breadth.”
“You need to be very skilled, i.e. have a high accuracy of being right, to implement a strategy effectively if there are only a limited number of investment possibilities. The problem with timing is that you only have handful of factors to select: If you time these factors you make a small number of decisions which can have a large impact on performance. This is not attractive from a risk reward perspective. We do allow for deviations from the strategically chosen factor mix, but these deviations are relatively small.”
"Instead of trying to time factors, we advocate a diversified approach. In addition to diversifying, investors should also have a long time horizon to absorb these periods of underperformance. This is not only the case for equities but also for credits.”
(8) How can factors be optimally mixed?
“The optimal mix is client-specific for two reasons. First, it should take into account your preferences. These preferences depend on the type of investor. Low-volatility stocks lose less in down markets resulting in a more stable funding ratio of pension funds while maintaining its exposure to the equity premium. On the other hand, value and momentum can be more attractive for sovereign wealth funds, because these funds are often more interested in maximizing returns. Second, it should take into account the portfolio’s current exposures in the integration of a factor solution.”
Stay informed on Quant investing
Receive our Robeco newsletter and be the first one to get the latest insights, or build the greenest portfolio.
(9) Do currencies have an effect when constructing a factor portfolio?
“Currencies can have an impact on total returns, but this is less than the factors themselves. They can also be hedged according to investor preference. There are many ways of dealing with currencies and we can create tailor-made solutions. We do not advocate factor strategies on currencies as a separate asset class, by applying a momentum strategy to currencies, for example.”
(10) Does a momentum strategy still work after you have taken the transaction costs into account?
“It is true that generic momentum strategies are inefficient because of the associated trading costs. Most research on momentum strategies has serious shortcomings and many studies do not take transaction costs into account. If you do, a considerable proportion of the returns disappear. Many generic strategies sell a stock as soon as it no longer falls within the top 10% of the universe.“
“However, it is possible to set up a more efficient momentum strategy and significantly reduce portfolio turnover by around 50%. In an efficient momentum strategy there is a tradeoff between the strength of the signal and the trading costs. And you should only engage in a trade if the signal is strong enough.”