“The most informed investors intentionally seek out opposing views and conflicting data sets to incorporate into their decision-making process.”
As a global community, people are struggling to access, process, and use data effectively to make good decisions. Bombarded by more information sources and more data than any previous generation, we must hone our abilities to curate diverse and relevant sources of data, intentionally sort through that information, and ultimately determine what is accurate and relevant. Only then can we act, if indeed action is necessary.
Consider the cacophony of voices recently commentating on market volatility. From Syntrinsic’s perspective, it’s a classic example of how incomplete data can inappropriately influence investor sentiment and potentially cause investors to make poor decisions. Although market volatility has been well within historic norms this summer and fall, throughout the media, at coffee shops and cocktail parties, and around investment committee conference tables, anxious comments and questions about volatility persist.
While we do not think that changes in market conditions over the past three months require a change in investment strategy, it is important to understand why so many investors perceive unusual volatility even though market movements have been typical.
Too often, investors start with opinions or emotions rather than with information. As the following data shows, volatility in equity markets has been well within historic norms while bond market volatility has been lower than historic averages.
As a reminder, volatility, in this case, is measured by standard deviation, which is defined as variability around a mean. In lay terms, that means you take the average of something over a period of time and see how much variation there is on BOTH sides of that average. The “BOTH” is key, since volatility can be positive as well as negative. For example, the US equity market surged from 1996 – 1999, then fell dramatically 2000 – 2002; however, the volatility measures during both periods are similar.
Data: Equity volatility has been typical
Understanding what volatility is statistically, consider equity markets since 1970:
- While US equity 12-month volatility through September 2019 is slightly higher than the long-term average (18.75 v. 15.04) it is consistent with many periods since 1970 when market returns were alternately positive, negative, and flat.
- US equity volatility was extremely low in 2017, the only calendar year on record in which the S&P 500 was positive every month. That consistency lowered volatility.
- In 2018, US equity market volatility spiked at both the beginning and the end of the year, causing 12-month volatility reported in 2019 to be more intense. The dramatic equity declines in 4Q 2018 were followed by significant increases in 1Q 2019. Since then, equity volatility has been fairly muted.
- Despite Brexit, the trade impasse between the US and China, geopolitical uncertainty, and political upheaval in many corners of the world, recent Foreign equity volatility is slightly below historic norms. In fact, recent volatility of the MSCI-All Country World (ex-US) Index is consistent with what the index has experienced since its inception in 1999.
Data: Bond market volatility has been low
While not as dramatic as equities, bond market volatility can reflect how investors feel about risk. Looking back to the mid-1970s:
- US bond 12-month volatility is quite low relative to the long-term average (3.73 v. 5.27).
- Weaving in data from Foreign Developed markets (where yields are lower to negative) and Emerging Markets (where yields are higher than in the US) the recent 12-month volatility also is lower than the historical average (4.32 v. 5.33).
- So, with all the uncertainty inherent in global markets due to central bank activity, slowing economic growth, geopolitical risk, country-specific risk (e.g. Argentina, China, etc.), bond market volatility remains quite low in both absolute and relative terms.
So, if equity market volatility is well within historic averages, and if bond market volatility is lower than historic averages, why then are so many people quick to express concerns about market volatility?
To answer that question, we must venture out of the land of data and head into the realm of behavior science where most investors actually reside.
Behavior: Investors confuse uncertainty and volatility
As we have established, volatility is a measure of variability around a mean, positive or negative. Uncertainty, however, is simply variability. Uncertainty can be variability around a norm (Can I trust that Denver will be sunny?), around an imagined norm (Can I trust the stock market to go up?), or around an expectation (Can I trust the health of the business environment?). Just as it is quite natural to wish for predictability when making personal and collective decisions, it is natural to want to rein in risk-taking when things feel more uncertain.
The less that we feel we can predict or trust future events, the more uncertainty we experience. And the more our personal uncertainty is magnified by the uncertainties of people in our networks, the greater distrust we feel. A problem arises when an otherwise perfectly reasonable human being starts to think that the uncertainty expressed in the newsfeeds, blog posts, and color commentaries they receive via their favorite sources means that they should not make long-term investments.
But this is nothing new. For a very long time, investors have built, bought into, or lent money to businesses despite economic and geopolitical uncertainty. What is different today?
Behavior: Humans struggle with gathering and using data
Over the past two decades of advising investors, we have seen a dramatic shift in how people receive, process, interpret, and act on data.
First off, investors must process more data than ever before in part because the companies that they are investing in are domiciled in and doing business all over the world. The policies, economies, and societies of countries such as India and Brazil matter because their consumers and their companies have become more relevant on the global stage.
Secondly, driven by need and enabled by technology, far more data is available far more quickly to far more people. The information provided in real-time by massive, broadly available databases (e.g. Bloomberg, FactSet, Bureau of Economic Analysis, etc.) gives ready access to information that was not available to anyone—not central bankers, analysts, or portfolio managers, let alone private investors—twenty years ago, certainly not in any kind of “real-time.”
All of this information (and misinformation) is then magnified by the little high-powered computers we carry around all day and much of the night. Our ability to access information and opinions continues to grow exponentially. While this increased transparency serves as an amazing planning resource for policymakers, businesses, and consumers, it also means that we see far more information than we ever did before and need to figure out how to manage that data flow.
Some manage data by relying on computer algorithms designed to filter and in some cases act upon this massive flow of data. For a hedge fund managing currencies, that kind of data interpretation and usage makes sense; however, in many cases, data consumers are not even aware of how much computer algorithms are filtering what information we are exposed to in the first place.
Once a data user has clicked on certain newsfeed articles or responded to certain posts or tweets, they have signaled what kinds of opinions they want to see. They have modified their data flow to attract information that confirms their bias. When we intentionally or inadvertently create data filters, we do not receive enough information that might provide a competing or balancing view.
Unfortunately, data and its interpretation are increasingly becoming a mark of tribal identity, a way that we signal where we get data, the types of opinions we have, and who we trust to inform us. It should not be as easy as it is to predict someone’s economic and market data sources by the opinions they express and the questions that they ask. We would wish to see more measured and balanced perspectives, which would require that people access more measured and balanced information.
Going forward wisely: Data + Behavior
The most informed investors with whom we work intentionally seek out opposing views and conflicting data sets to incorporate into their decision-making process. At a time when many people only want to talk to people who agree with them, thoughtful investors seek out people with diverse opinions in order to test their theories and strategies. Syntrinsic’s internal investment committee makes our forecast, asset allocation, and investment manager decisions by consensus precisely so that we can hash out the implications of diverse perspectives, challenge and justify our views, and emerge with more effective decisions.
As it relates to volatility, we recommend that investors hold steady, assuming their portfolio aligns with their objectives and risk profile. As it relates to information gathering, we recommend that investors intentionally expose themselves to diverse perspectives that enable them to make well-informed decisions. We promise to continue doing the same.