In dealmaking, detecting cognitive bias—those autopilot thoughts that have a habit of sending us down familiar pathways—feels a bit like particle physics. Their presence is undeniable, but proof is harder to come by. That’s because biases (like subatomic particles) have a tendency to oscillate at nearly undetectable levels.
They’re hidden in outcomes and decisions and not overt in the decision-making process. And when they do appear to be influencing decisions (like valuations, for instance) it can be hard to isolate where they took hold.
A Change of Heart and Numbers: The Snap IPO
Take Morgan Stanley’s double-take on the Snap IPO in 2017. A few weeks after helping the messaging app go public with one of the most high-profile tech IPOs at the time, the investment banker (and lead underwriter on the IPO) issued its first research note, bestowing a “buy” alongside a $28 price target for the social media company’s shares.
Within 24 hours, Morgan Stanley issued a second note, reportedly stating that they had “corrected a tax calculation error in our model that overstated adjusted EBITDA in 2021-2025.”
The investment bank had kept the revenue forecast and fundamental top-line drivers like daily active users and ad loads the same, but had done a serious re-work on the app’s adjusted EBITDA from 2021 to 2025, a change which hacked $1.7 billion in estimated adjusted EBITDA in 2025 alone. And yet, they landed on the same price target: $28 per share.
Surrounding the change, Charles Lee, a professor at the Stanford Graduate School of Business, told Business Insider it almost felt like the investment bank was “backing into the numbers” with the updated report.
“It’s almost humorous,” Lee told Business Insider. “And, of course, it can be totally innocuous, and it just so happens they found two offsetting errors, and that’s their opinion, and the price target should be unchanged. One has to sort of chuckle because there seems to be so much play in the numbers that they could have put anything in.”
The Varieties of Bias
So, was it confirmation bias: favoring information that confirms your previously existing beliefs? Or maybe affiliation bias: the inclination for analysts to be overly optimistic when their employers have underwriting relationships with the firms they’re covering? How about pro-innovation bias: excessive optimism and a greater value placed on the usefulness of a particular innovation in the future without considering its weaknesses? All are plausible, but hard to prove. That’s because cognitive biases are ingrained. They shove us in certain directions.
Valuation isn’t a science. There are variables and assumptions and adjustments. There are future concerns and predictions made. And for dealmakers, it’s important to understand the types of cognitive biases that can sneak into the valuation process.
Confirmation bias, a predilection to favor information that confirms previously existing beliefs, is a well-trodden cognitive bias. It’s also versatile. According to an article in Psychology Today, confirmation bias can take the form of preference for evidence supporting beliefs, or interpreting data in a way that is in line with that assumption.
But it can also mean dismissing or ignoring information that might test those beliefs. “The bias degrades our judgments when our initial beliefs are wrong because we might fail to discover what is really happening until it is too late,” writes Gary Klein, the article’s author.
In the valuation process, confirmation bias can influence the way certain data is treated, according to the study “What Drives Merger Decision Making Behavior?” published in the Journal of Economic Behavior and Organization by Vicki Bogan and David. “Within the context of a merger, cost information may be thought to have greater potential to disconfirm the merger, while information on savings may more likely confirm the merger,” write the authors. “Thus, we expect individuals will regard cost and savings information very differently depending on their prior inclination toward the merger.”
Get TOUCHPOINT stories delivered straight to your inbox.
As Stanford’s Lee explained to Business Insider, Morgan Stanley’s direct stake in the success of Snap can’t be overlooked. “If you’re an investor, anybody who really cares about the long-run value of this bet, you probably want to discount this (their report) more than the others, given their affiliation (as the IPO underwriter).”
Research has examined the role of affiliation bias in valuation, especially when it comes to IPOs. It’s an often unavoidable conflict of interest that can arise, especially given that the financial institutions provide both analyst research and investment banking services.
“Underlying this conflict is the idea that analysts provide optimistic research coverage in an attempt to curry favor with their firm’s existing clients or to win future investment banking business from covered firms,” writes Stephannie Larocque in a piece for the Harvard Law School Forum on Corporate Governance. She points out that while efforts and regulations have been made to separate the two sides of the business for sanctioned banks, these industry-wide SRO rule changes were “largely ineffective at reducing the influence of investment banking on analyst research” at non-sanctioned banks. Affiliation bias, it seems, is a hard system to rewire.
The past decade has seen an increasing abstraction of tech company valuations. As of March 2020, there are 451 tech unicorns (companies valued at more than $1 billion) according to CB Insights. They’re household names like Fiverr, Slack, Uber, Zoom and Pinterest. And yet many aren’t profitable and are trading lower than their debut price. According to an article in Vox, 81 percent of US companies in 2018 were unprofitable the year before they went public.
“The rise in unprofitable IPOs reflects the general preference in both public and private markets for growth over profitability,” Paul Condra, Pitchbook’s lead analyst of emerging technologies told Recode. “As we’ve seen during most of the recovery period since the Great Recession, investors are not so margin-focused, but continue to put a premium on businesses with long-term future expansion or disruption potential.”
Part of that stems from pro-innovation bias, excessive optimism and value placed on the usefulness of a particular innovation in the future without considering its weaknesses. The value is a reflection of this hope that future profits will eclipse current losses. But things change, especially in technology.
Snap offers a cautionary tale for how getting caught up in assumptions and biases can ripple out further down the line. In February, the company admitted it agreed to a $187.5 million settlement in a class-action lawsuit where investors alleged that the social media company had “understated the threat of Instagram.” There’s a lot of ways to read that, but the numbers don’t lie. Since its valuation and IPO, Snap has never come close to hitting that $28 target again.
Illustration by Christy Lundy