Justin Sibears is a Managing Director at Newfound Research. He’s responsible for the ongoing research and development of new intellectual property and strategies. Justin also serves as a Portfolio Manager for Newfound’s direct offerings. We recently spoke with Justin, who will be speaking at our ETF Strategy Summit (Oct. 15 – 16 – Dallas), as he shared with us how they incorporate behavioral finance into their investment process.
ETF Strategy Summit: Factors, geographies, sectors, asset classes, styles, strategies – given today’s environment, are there any methods for slicing up the markets with ETFs that are clearly superior to others?
Justin Sibears: We believe it is equally important to view a portfolio through all of these lenses. We have a saying at Newfound that “risk cannot be destroyed, only transformed.” Oftentimes, viewing risk through only one of these lenses may lead us to make decisions that appear to reduce risk, but instead simply shifts the risk in a way that hides it from view.
As example, consider value strategies. Value strategies can be classified in many ways, but one important way is how they manage sector risk. Some value strategies are sector unconstrained. They will pick the cheapest stocks regardless of what sector they are in. Some value strategies constrain their sector allocations to be in line with the broader market. Others may fall somewhere in the middle. It’s entirely possible that two value strategies, one sector unconstrained and one constrained, may quantitatively have similar exposure to the value factor. If we stopped our analysis here, we may miss potential sector exposures that could be problematic. For example, the investor may have a large strategic position in energy stocks and so may not want to own a value strategy that can simultaneously make a large bet on energy.
ETF Strategy Summit: What methods can advisors use to keep abreast of a portfolio’s exposures as they blend things like style, factor and active ETFs with alternative strategies, country ETFs or sector ETFs?
Justin Sibears: We recommend that advisors blend quantitative and qualitative tools. Today, there are many quantitative tools that allow advisors to measure risk and stress test portfolios. These are undoubtedly valuable. However, it’s also important to note that these tools have limitations. They are only as good as the assumptions they make. Relying on quantitative tools exclusively has the potential to lead to over optimization and false confidence. To protect against this, we advocate complementing quantitative analysis with simple qualitative rules-of-thumb. As an example, we think it’s hard to go wrong with equal-weighting exposures, whether that be sectors, factors, or asset classes. As Harry Markowitz, the father of modern portfolio theory, said, “I should have computed the historical co-variances of the asset classes and drawn an efficient frontier. Instead, I visualized my grief if the stock market went way up and I wasn’t in it – or if it went way down and I was completely in it. My intention was to minimize my future regret. So I split my contributions 50/50 between bonds and equities.”
ETF Strategy Summit: You’ve noted that modern portfolio theory is imperfect – how are you incorporating client behavior to optimize asset allocation?
Justin Sibears: At a high level, we incorporate behavioral finance into our investment process by acknowledging that investors dislike losing money more than they like making money. We know that investors care deeply about protecting the capital they have worked hard to accumulate and as a result our first focus is on risk. Specifically, the majority of our strategies seek to improve risk-adjusted returns and manage sequence risk by avoiding large drawdowns.
We also recognize that while investing shouldn’t be complicated, that doesn’t mean that it is easy. In the real world, short-term emotional decisions can threaten even the best-scripted financial plan. As a result, we believe strongly that the optimal investment strategy for any client is, first and foremost, one that they can stick with. This means recognizing that managing client anxiety is paramount for long-term success.
We take a multi-dimensional view of anxiety. Specifically, our optimization process is built around the following three tenets:
- Not risk averse, but loss averse. In modern portfolio theory, volatility is used as the main metric of risk. Investors, however, are not necessarily risk averse. In fact, a 2014 paper by Frazzinni and Pederson finds that investors actually have a preference for high risk, “lottery” style investments. By using volatility as the primary measure of anxiety, modern portfolio theory punishes both bad, downside risk and preferable, upside risk equally. Our process focuses explicitly on downside risk.
- Keeping up with the Joneses. When evaluating outcomes, humans often have an established reference point. Outcomes are classified as gains if they are above the reference point and losses if they are below. In investing, reference points are often established public benchmarks (e.g. the S&P 500), but may also be the performance of peers. Our process considers tracking error as one measure of anxiety. If we are going to make bets that move the portfolio away from the benchmark, we have to be adequately compensated as these bets will inevitably lead to angst during periods of short-term underperformance.
- A preference for a smooth ride. While modern portfolio theory is concerned with optimizing for the end result, investors live in a continuous environment. Behavioral research finds that investors who monitor their portfolios more frequently will actually perceive their investments to be riskier: a phenomenon known as myopic loss aversion. Framing investment results over short investment horizons, compared with an asymmetric view of gains and losses, creates a preference for a “smoother” ride over time. As a result, our process is willing to sacrifice some long-term return for a more enjoyable experience since this can reduce investor anxiety and increase commitment to the investment plan.
ETF Strategy Summit: You’ve also advocated for lowering withdrawal rates as the assumed future rates of return for traditional investments decline – should investors and advisors also become more comfortable with risk taking to achieve their goals?
Justin Sibears: It’s hard to make a credible case for future stock and bond returns living up to past experience. The unfortunate consequence is that financial rules of thumb – like the 4% withdrawal rule – may fail going forward since they were calibrated to the past. One easy solution for retirees is just to withdraw less money. For obvious reasons, this solution is far from ideal, if not impossible for some.
There is no silver bullet for solving this problem. Fortunately, we do believe that there are many marginal improvements that when compounded can get withdrawals back up to acceptable levels. Some of these improvements are more planning than investment centric (e.g. managing fees, increasing savings, being smart about asset location, and using dynamic withdrawal strategies).
On the investment front, we do believe that risk plays an important role. However, it’s not as simple as being comfortable with more risk. When expected returns are high across the board, we have the luxury of making risk profile choices as a matter of personal preference. If we are a conservative person, we can allocate conservatively and vice versa. In our current environment, however, this is no longer an option for many people. Instead, it’s crucial to choose a risk profile that has a realistic shot of delivering on a client’s long-term objective even if this means taking more risk than you might otherwise take in a perfect world.
We also believe that it’s very important to diversify your diversifiers. Low interest rates are so detrimental to retirees because they make the most common risk management tool, bonds, extremely expensive to hold in large quantities. In this type of environment, one way to improve returns is to move beyond fixed income as a risk management tool. This means incorporating other asset classes and strategies that have a proven track record of strong performance in times of financial crisis. One example of an investment style that fits this profile is trend-following.
ETF Strategy Summit: How are you using active and passive products together in a meaningful way in your portfolios?
Justin Sibears: We use active strategies that attempt to harvest the risk-adjusted outperformance of evidence-based factors like momentum, value, carry, and defensive. Generally speaking, these types of strategies can outperform over the long-run for three reasons. First, the outperformance may be compensation for bearing risk. Second, the outperformance may be the result of taking advantage of other investors’ behavioral biases. Third, the outperformance may result from taking advantage of structural inefficiencies in the market.
In all three of these cases, underperformance, especially over the short-run, can and will occur. This underperformance is a necessity for the premium to exist. If it never occurred (i.e. if a premium is compensation for bearing risk and the risk never materializes), too many investors will adopt the strategy, capital inflows will drive up the prices of the underlying securities, and the forward return of the strategy will approach zero.
As an example, investors are broadly aware that value investing has historically generated excess risk-adjusted returns. However, actually capturing this return required suffering through excruciatingly long periods of underperformance (consider that the Barron’s cover article in December 1999, What’s Wrong, Warren?, opined that “Warren Buffet may be losing his magic touch.” The bubble peaked three months later.).
Simply: if we expect to generate long-term outperformance, we must expect periods of potentially short-term underperformance in which weak hands that fold pass the premium to the strong hands that hold. The trick of asset allocation, in our mind, is to make the ride bearable so that investors stand a chance of staying committed to these active strategies so that the benefits can be reaped over the long-term.
We seek to achieve this goal in two ways. First, we seek to diversify both across active strategies (e.g. pairing momentum and value) and within active strategies (e.g. diversifying how our value strategies measure value, think P/E vs. P/B vs. EV/EBITDA). Second, we recognize that this type of diversification may reduce the magnitude and duration of underperformance, but it will not eliminate it. Therefore, we blend passive strategies into the portfolio to manage tracking error. In addition to helping us manage tracking error, the passive component allows us to manage overall portfolio cost to a level that we deem acceptable.
ETF Strategy Summit: Thanks Justin. We look forward to hearing more of your thoughts at the ETF Strategy Summit October 15 – 16 in Dallas.