Behavioral Biases Series: Recency Bias
In my opinion, this week’s bias is one of the easiest to comprehend, but one of the hardest to avoid. Recency bias is classified as a cognitive, as opposed to emotional, bias, and specifically pertains to an information processing error that stems from the recency effect.
In psychology, the recency effect is the phenomenon that recent occurrences are more likely to persist in short-term memory than previously ‘older’ happenings. Put simply, you can recall far easier what you had for lunch today than what you had for lunch on this same day last year. Thus, recency bias pertains to errors we make as a result of allowing the recency effect to prejudice our recollection of the past, and unduly influence our decision-making in the present.
In probability terms, recency bias causes humans to estimate future probabilities based on only a handful of recent outcomes instead of long-term experience. Within investing, overweighting the occurrences of the most recent past can be a slippery slope to underperformance and often fuels trend following: where investors pile into holdings that are doing well and sell out of holdings doing poorly as of late. As these decisions usually “feel right at the time,” it is important to remember that within investing (not speculative trading), 1-3 years is a short period, and will not usually capture the cyclical nature of asset class returns. To the naïve investor, high performing asset classes of the past 1-3 years may seem unduly attractive. In general, one should be more skeptical to purchase the most recent winners, as they will seldom reclaim that status the following period. For more on why, refer to our past article on mean reversion.
To illustrate how overweighting recent performance when making decisions can be detrimental, let’s have a look at a 20 year “Periodic Table of Returns” generated by BlackRock:
Each square represents a different index, with the white box representing a diversified portfolio comprised of multiple asset classes in a classic “60/40” portfolio, with 60% equities and 40% fixed income. Notice how the white box is consistently in the middle range of performance, and in the top third on the list of annualized returns over the 20 years. In contrast, notice the performance of the S&P commodities index (yellow), or the Russell 2000 Growth (brown), they are great examples of how the recent past (1-3 years) should be properly contextualized, and not over-extrapolated into the future for the simple reason that they’ve recently had success. Likewise, the same should be said for single year worst performers, as investors will often avoid them for the same reason: undue expectations that recent (poor) performance will surely persist. In any given year, an individual index can have a very different performance than its recent past or historical “normal” level, hence, the higher level of stability in performance of the diversified portfolio.
An investor suffering from recency bias would most likely make their investment decisions solely based on the past 1-3-year performance. This behavior has come to be known as “strategy chasing,” and often resembles a counterintuitive practice of buying high and selling low.
Ask yourself, have you suffered from recency bias, perhaps while choosing from a list of funds in your company’s 401k? If so, you are not alone, and you may have experienced this phenomenon first-hand, although hopefully not to an extreme. It is human nature to be immediately attracted to the perceived winners and immediately repulsed by the perceived losers – but don’t be too quick to judge – have another look at the Russell 2000 Growth box between 1999-2003. Talk about a wild ride! In hindsight, however, it is all but one example of countless times where recency bias would historically lead investors astray.
Although the recent past may be freshest in your memory, when making investment decisions you should always put performance statistics or recent occurrences into proper context as it pertains to your own investment horizon and objectives. To avoid letting the recent past have an undue influence on your immediate decisions, recency bias can be apprehended by 1) using large data sets for analysis, spanning multiple economic cycles 2) creating a rules-based system for making decisions 3) maintaining proper diversification, and 4) a healthy skepticism to the little voice in your head saying, “times have changed…this time is different!”
Samuel Frymier is a co-founder and managing partner of Quesne Capital. Sam earned a Bachelor of Science in Finance from Duquesne University, graduating in the top 10% of his class. Once a nationally recognized division one soccer player, Sam now coaches at the youth level. Outside of soccer he has always had a love for golf.[button label=”Get The Latest Blog Posts and Podcasts Right In Your Inbox!!” link=”#ppsShowPopUp_106″ target=”_self” color=”blue” __fw_editor_shortcodes_id=”6e61af2aad0f6ed11772d171cd0e945e” _fw_coder=”aggressive”][/button]
Disclaimer: This information is for educational purposes only and should not be construed as investment advice. Please refer to our full Disclaimer for more information