Q: How has the fund evolved since inception?
We wanted to design a product that systematically exploits the low-volatility anomaly, but also carries substantially less risk than equity investing. That’s how we developed the idea of building a long/short portfolio. The fund has a track record of four to five years, but the strategy, which focuses on low volatility or low-beta equities, has existed since 2010.
For every $100 the client invests, we have a long portfolio of $100 in stocks with beta of less than one, as well as a short portfolio of $30 in stocks with substantially higher beta. Over the long run, low-beta stocks outperform high-beta stocks, so there is an inherent advantage in the long portfolio, which systematically has higher alpha than the short portfolio.
The idea behind the strategy is that by picking shorts out of high-beta stocks with poor fundamental characteristics, we are able to hedge the portfolio. Our goal is to deliver equity-like returns with considerably less volatility. I believe that our portfolio construction technique is fairly unique.
Q: What is the priority in your strategy? Is it lower volatility or equity-like returns?
The standard deviation or beta is easier to control than the expected return. We specifically target the risk, so risk is always maintained, while the return is something that we hope to achieve over the full market cycle.
Q: How would you define a typical market cycle? How many years does it last?
For me, the market cycle is the environment where the start and end points are the same. In other words, a market cycle usually includes a period when everybody runs away from risk, followed by a period of high volatility. Then the risk abates and risk aversion is back to its original point. It can happen in a year or it can take 10 years; it is difficult to define it in a time period.
Q: What are the core beliefs behind your investment philosophy?
I believe that the investment process needs to be systematic; it needs to follow a defined set of rules. The systematic approach prevents the process from being influenced by emotion or investor overreaction. For us it is important to have a set of written rules and to follow them.
Second, we are very clear about the source of our return. We target equity-like returns; we are long $100 and short $30, but we are 70% net long. So, the first component of return comes from the equity market risk premium. The second component of return comes from the low-volatility anomaly. The third component of return comes from our ability to exploit the characteristics that have historically been rewarded in the market, such as value and quality, because they continue to get rewarded.
Overall, our investment philosophy is based on being systematic, following written rules, and believing that low-volatility stocks will beat high-volatility stocks in the long run.
Q: What is your investment universe? Do you have any preference for certain geographic regions or sectors of the economy?
Our view is that the wider the universe, the better, as long as we can buy and sell stocks at a reasonable cost. Our universe represents the largest 1,800 stocks in the world, or the stocks in the MSCI World Index, because we have a minimum liquidity threshold. We’ll go long and short across different countries and sectors.
Theoretically, the stocks can be anywhere in the world as long as they are in developed markets. So, we can invest in North America, South America, Europe or Asia, but not in the emerging markets. The logic behind this approach is that if we go long in a developed market and short in an emerging market, we would face a different kind of tilt, which is not part of our investment process.
Q: Could you describe your investment process? Where does it start?
Our process has three critical steps. We start with two different ranking systems, which we apply to the universe of 1,800 stocks. The first and more important system is risk assessment, because the fundamental component of our process is to go long in low-risk stocks and short in high-risk stocks. So, we rank the stocks based on their riskiness by using two different risk models.
One of them is a fundamental risk model, which allows us to evaluate stocks based on their country or industry. However, there are stocks that don’t neatly fall into a single industry group, especially with the way the economy has evolved over the last three years. For example, Amazon can be considered consumer staples, retail, or cloud computing. To overcome that hurdle and to calculate the risk of a stock, we also look at the stock’s market correlation from a price return standpoint.
We use both of the measures to categorize stocks into low beta or high beta. When we say that a stock is high beta, it’s not just from a fundamental standpoint, but also from the way it behaves. So, the first part of the process is to sift through the universe to figure out in which stock to go long and in which one to go short.
The second component of our process is the alpha engine. It takes every stock and looks at its fundamental characteristics. We examine over 50 different fundamental characteristics, such as price/earnings levels, trailing P/E and its acceleration, estimate revisions, etc. After we collect all the data, then our expertise comes up with the weight we place on each of the characteristics. When we know the weight of each characteristic, we use our proprietary technology to determine a score for each stock.
In other words, if we are early in the market cycle or economic recovery, valuation tends to matter. At a later stage of the economic recovery, quality and profitability become more important. The goal is to identify the factors that are in favor, based on factor momentum or factor mean reversion. This part of the process results in ranking all the stocks in the universe from best to worst. Basically, we perform risk assessment for each stock. Based on this assessment, we decide if the stock will go in the long or the short portfolio universe. Then, within each universe, we rank all the stocks.
The next step is constructing the portfolio to meet our overall risk goals of being long 100 and short 30, while having a beta of about 0.5, or half the risk of the market. One of our tools is a portfolio optimizer, which evaluates the transaction cost not only of buying the stock, but also of selling it.
We also have a process to screen out unusual events. For example, the company may be involved in a takeover, regulatory investigation or Board investigation. We take these events into account when we construct the portfolio. We also look at the recent news on the stock and the implied volatility as a way of estimating the forward-looking risk. The company’s ESG scores are also important. Companies with poor ESG scores tend to have unexpectedly large stock price movements, we take small positions in those companies. We embed all of these constraints into the optimization process.
We have a systematic way of ranking all the stocks from a desirability perspective on a daily basis. Because things don’t change too much on a daily basis, we typically have 10% to 20% turnover every month.
Q: Do you group the fundamental characteristics in different categories?
We don’t put the factors into groups for the analysis, but we group them for attribution. We use about 10 different valuation metrics and various growth measures, which show how fast the company is growing. We also use measures of liquidity in terms of the current volume and the volume relative to the stock’s long run average, as well as relative to others in the industry. Another set of measures is related to financial risks and the earnings quality. It evaluates if earnings come from cash earnings or from accruals, if the company is highly profitable, what’s the asset turnover, etc. All of these measures are covered using a variety of different characteristics.
Q: What is your approach to the tactical weight allocation?
We have identified two systematic factors in terms of market rewards. First, the market tends to reward factors on a persistent basis. If a factor worked last month, it tends to work this month. And if a factor worked last year, it tends to work this year. So, there tends to be factor persistence on a three-year basis, because the last one to three years are very informative in terms of the factors, which are going to work going forward. That’s what we mean by factor momentum. The second informative aspect is mean reversion in factors, and I think all factor returns come from investor behavior.
Q: How different is your strategy regarding the short portfolio? What additional work goes into it?
Shorting is not that different from going long, but there are two main differentiators. The first one is that stock returns are not normal. Stocks go up more than they go down. When we budget the risk for a stock, we aim to limit potential losses, so we would allocate a larger position to a low-beta stock. However, accounting for the fact that stocks go up more than they go down, we need to make sure that our short book is more diversified than the long book in terms of individual stock concentration.
The second and more important rule is related to human behavior, specifically to the difficulty admitting a mistake. When you are long and make a mistake on the stock, the mistake is self-corrected. In a systematic process like ours, where the risk controls would signal exceeding the risk in the stock, we would trim down the position. So, in the long book a mistake is not that critical, because it would correct itself. In the short book, however, when we are wrong and the position has become bigger than it should be, we have to bring it down in a way that’s appropriate to the risk of the stock.
Q: Would you elaborate on your optimization process?
The optimization process really allows us to build a portfolio that is long $100 and short $30, or to achieve $70 of exposure with beta of 0.5. The transaction cost is an important factor and is incorporated into the optimization. We focus on minimizing transaction costs and that’s a huge component of our investment focus. Our earlier allocation of resources used to be just portfolio management and research, while now it is portfolio management, trading and research. We believe that trading is really important as a way of saving costs and boosting performance.
Q: What is your portfolio construction process?
We spend considerable effort on forecasting the volatility of each stock. We look at the trailing volatility of a stock, the implied volatility from the options market, the ESG score and the news. We use all that data to adjust the forecast volatility of the stock, which in turn affects the relative position size
Our portfolio limits include a rule about maximum position size, so if a stock moves three times its standard deviation, it shouldn’t cost us more than 50 basis points. Regardless of the volatility, though, we can never take more than 3% in a position. That’s what determines individual position size.
We also have a set of rules around diversification in terms of the maximum exposure to any single industry, country, or region. These rules ensure that the portfolio has broad exposure to a wide variety of sectors and countries.
Q: How do you select stocks when the markets are sporadically volatile?
We look at three components. The long equity portfolio generates return in a slightly rising market, while our short book produces value in a falling market. The question is what to do when the market isn’t going anywhere. That’s where we have a stock selection process with a tilt to over 70 different factors based on their recent tendency. Currently, we place huge weight on quality and recent profitability, so if the market or high volatile stocks are going nowhere, these are the factors that will deliver returns for our clients over the next month.
Q: How do you define and manage risk?
We target beta of 0.5 and that means that if the market is down, we have 50% less downside volatility than the cap-weighted market index. We always maintain a consistent risk profile. We are explicitly not varying the risk profile of this portfolio; it is always 100 long, 30 short and always has a beta of 0.5.
Our goal is not to time the market, but to always have a short portfolio of 30% of high-beta stocks. We try to add value by shorting high-beta stocks that underperform in the long run. We get additional return by overweighting and underweighting the factors that investors are paying attention to, such as quality, value, leverage, etc.
The 2000–2002 period was an inspiration for this strategy, because it showed how weak the models were in terms of identifying the difficult investment climate. I thought that there had to be something better to rely on for portfolio protection, something that will always be in place. I believe that the basic principle for building any investment portfolio should be the underlying thinking of what’s going to be in place to help during a tough market environment.
Living through 2001 was really the inspiration that led me to the idea of low-volatility equity investing, which we started doing in 2004.
