There is a psychological phenomena called the Dunning-Kruger effect. It states that unskilled people may not know enough to figure out that they are below average. Dunning et al recently wrote in Why People Fail to Recognize their Own Incompetence (Current Directions in Psychological Science, 6/23/16):
…people tend to be blissfully unaware of their incompetence. This lack of awareness arises because poor performers are doubly cursed: Their lack of skill deprives them not only of the ability to produce correct responses, but also of the expertise necessary to surmise that they are not producing them.
It may feel good to think about all those “unskilled people” who go through their lives blissfully confident and unaware that they are unskilled, but to do so misses the point. By definition, half of us are below average and all of us are unskilled at some tasks.
The Dunning-Kruger effect is closely related to the overconfidence bias and self-serving bias. The overconfidence bias is the tendency for humans to be over confident about their skills, abilities and choices. Self-serving bias is when we attribute success to our skills, and failures to external factors or other people.
Unfortunately, overconfidence can be the most destructive bias and is one of the most pervasive. Everybody is overconfident, even you and me. Daniel Kahneman who wrote the book about overconfidence said he believes he is still overconfident. Self-service bias helps protect our self-esteem. Frank Knight stated that sometimes it is the belief in our own luck, and this bias dates back to the bible when Adam blames Eve for his mistake in eating the apple. Self-serving bias can be witnessed daily in the Wall Street conference call.
I cover these biases in my class on Behavioral Finance at Golden Gate University. Given a stack of behavioral finance books and some time on the internet you will find that humans aren’t the rational beings we think we are and are frequently biased in certain directions. The increasingly detailed definitions of biases seem to be primarily about generating academic papers, but the Dunning-Kruger effect is worth thinking about because it happens a lot in business.
B
usiness leaders want to appear upbeat, positive and confident, and acting this way (see here) tends to make us confident. This isn’t the same as “fake it to you make it” (which I recommend to my students) as you know when you are faking it. Kahneman stated in his book Thinking, Fast and Slow that “An unbiased appreciation of uncertainty is a cornerstone of rationality— but it is not what people and organizations want.” We want certainty in our leaders, not complexity. There are no clear tests of business ability so it is easy for executives to be overconfident of their skills.
Ignorance more frequently begets confidence than does knowledge: it is those who know little, not those who know much, who so positively assert that this or that problem will never be solved by science. Charles Darwin
Hiring
I used to think I was a good picker of executive talent. Of the five controllers I’ve hired, the majority have gone on to be CFO’s, and I’ve had successful hires for CIO’s, President, VP Human Resources amongst others. I recently talked to Dave Arnold, a recruiter who has hired hundreds of CFO’s with success. Dave interviews 20-30 candidates a week and used to teach interviewing skills to management teams. My confidence in my interviewing skill was based on a small sample size and a comparison to other executives who may not have been great interviewers either. There is a lot of evidence that interviews aren’t that helpful for many jobs, due to interviewer incompetence and candidate dishonesty (see here and here).
I once worked with a leading recruiter for a COO position who kept sending us candidates who lacked analytical skills. Eventually after a long search the CEO hired someone. Within a year I was walking the new hire through their termination paperwork. Conversations with the recruiter revealed that the CEO had been very specific and the “weak” candidates were exactly what was requested. The recruiter knew that the CEO was wrong but eventually caved in. CEO’s who express confident answers even while they lack knowledge is known in Silicon Valley as “Founderitis” or Founder’s Syndrome (see here and here) another form of overconfidence.
With a small sample size, you are especially prone to be overconfident about hiring skill. All failures are blamed on the candidate, all success is attributed to management competence. Hiring a competent recruiting partner lowers risk and makes you smarter.
Investing
Private Equity fund managers are quite confident at investing in growth companies, even when evidence doesn’t support their confidence or skill. As a hedge fund analyst, I followed firms who kept growing even as signs of declining productivity and performance became overwhelming. Rue21 in 2012 boasted that they were opening units in Paris (Texas), London (Kentucky), even while average volumes in new units were down double digits. Apax Partners then paid $1.1b to buy this 877 unit chain in early 2013 and grew it to over 1,200 stores by the end of 2016. Associated Press reports that Rue 21 is closing 1/3 of their units in 2017 to end back at ~800 stores. To spend four years to end up where you started is a bad outcome. The trends in the industry (e-commerce!) and in Rue’s numbers (declining ROI) have continued to the surprise of their investors. And unfortunately, there is a significant chance this isn’t the last bad news out of Rue.
During the 1980s there were a group of auditors out of Texas that used to celebrate the completion of a savings & loan audit with a party which featured drinking champagne out of their boots. (I didn’t witness this, darn! I heard the story from an audit partner who’d paid a sizeable sum of money to settle the claims.) 
A couple of things you should know. First, you ruin good champagne by pouring it in footwear, and secondly, wine is not particularly good for leather. Auditors who think this is a good idea probably aren’t appropriately risk and control oriented. And investors in businesses that need risk management (like a S&L) shouldn’t hire champagne swilling auditors who can’t afford stemware. The person who told the story considered himself a savvy businessman. He put the blame for the eventual failure of the S&L and expensive legal claim down to bad management, the economy and regulators and none to the audit team or the firms’ lack of risk management.
Elizabeth Holmes dropped out of college to start Theranos at 19 with $6m in venture money. This ballooned into an eventual $700m of private equity/mutual fund financing. Ms. Holmes reportedly used to keep the office in the mid-60s so she could wear her signature outfit: black mock turtleneck, black pants and puffy black vest. In January 2017, Theranos laid of 40% of the staff and in April 2017 agreed to leave the blood testing business for two years to avoid further sanctions by the Centers for Medicare & Medicaid Services.
Who invests in a company with a CEO who can’t figure out how to take off her vest? Although Ms. Holmes wanted to be Steve Jobs, she didn’t get that even Steve Jobs wasn’t always Steve Jobs.
Nassim Taleb talks about Black Swan’s as unpredictable or unforeseeable events. I’ve commented in the past (here) that “black swan” events decline as learning increases. In this case, it appears that Silicon Valley VC’s didn’t participate in the funding of Theranos and most of the money was private equity and east coast VC funds. Vanity Fair reported that Google Ventures staff had attempted to get a blood test using the proprietary Theranos Edison machine, and found instead of taking a pin prick, the test used the same samples used at any clinic. Google Ventures passed on the investment. What did they know that $700m in capital and a board full of luminaries did not? Why no medical expertise on the Board? Why was there no CFO?
Theranos was intensely secretive about its Edison technology which, based on the latest disclosures by the company, never really worked. Probably smart keeping it a secret. Every investment mistake isn’t due to overconfidence, of course, but these examples show that sometimes even the experienced are unaware of how unskilled they are.
Human Resources
There has been a litany of “bad behavior” stories in the paper recently with Uber, Snapchat, GitHub all being hit for sexual harassment and not following basic human resource standards (see here). These companies have sufficient money to hire a senior HR executive but they didn’t think they needed one. That was a bad choice. Between the bad publicity, the lawsuits and the costs to correct the problems, these firms and executives are learning the cost of ignorance.
Scaling a rapidly growing business isn’t easy. Not all of it is renting space and hiring. Being really good at technology or raising money or even having a terrific idea is helpful but it isn’t enough. It takes a team of executives. We shouldn’t be surprised however when we hear of these problems, Uber won’t admit that it has a CFO (see here) and Tesla and Github are both going through relatively rapid CFO changes (see here and here). The CEO’s including Travis Kalanick (Uber), Elizabeth Holmes (Theranos) and Chris Wanstrath (Github) had no experience running any successful business prior to their startup.
Learning is at least partially about reducing unforeseen, unexpected and negative events. There are techniques for overcoming our overconfidence and Dunning-Kruger (see here, here and here) and if practiced, can help. We also can be amused at the unskilled and how they don’t realize how incompetent they are. But we shouldn’t be amused without realizing that sometimes we are the unaware and unskilled.
***
Dr. John Zott is the Principal consultant at Bates Creek Consulting. John is the chair of the Careers Committee at FEI Silicon Valley, a senior adjunct professor at Golden Gate University and comments regularly on issues that affect consumer businesses. If you are looking for a CFO for your e-commerce/retail/consumer company, or are a former student, colleague or would just like to connect – reach out.