Entrepreneurial Escalation Bias

How can such a smart investor make such a bad choice… again?!

A while back, I evaluated the viability of a major manufacturing unit and realized that the business unit was not going to be profitable. In spite of many millions of dollars invested and the fact that the 3 year project hadn’t even been commissioned yet, I had to recommend that the unit be sold off, or the product line changed dramatically. It wasn’t the last time that a business leader became angry with me for stepping on their commitment bias. But inevitably, the numbers caught up to them. A year later, the investment collapsed, and they sold it off.

In Accounting Psychology we address the biases and heuristics that go into financial judgement and decision making. One of the most powerful biases I have seen in the market is called the, “escalation of commitment,” bias. This bias keeps an investor attached to a losing investment simply due to investment momentum. The investor cannot admit to themselves that an investment is bad, and rather than allowing the investment to die out, invests more, even though there may be no indication that the investment will perform any better. It’s so dangerous because the investor becomes increasingly blind to the point where they can no longer see or hear helpful advice.

The escalation of commitment bias is driven by two cognitive processes:

1.      Justification: the story that the investor tells him/herself about why the investment is reasonable in spite of indications otherwise, and,

2.      Sunk Cost: the amount of money invested. The more there is invested, the greater the cost of walking away, thus commitment grows as the sunk cost grows.

This bias is measured by two key factors:

1.      Escalation Likelihood: The degree to which an investor is likely to invest more.

2.      Increment Magnitude: The amount of money the investor is likely to invest at each increment based on past investments.

The research shows that likelihood is increased by repetition: i.e. the more times they invest, the more times they are likely to invest. Also, the size of the investment is likely to go up each time they invest.

Now for the really scary part… this bias is especially true of entrepreneurs. Markovitch, et al. (2014) showed that in groups, people who described themselves as “entrepreneurial” were more likely than average to escalate their investment commitment, AND more likely to invest more money per increment. Perhaps this is because entrepreneurs tend to have a greater tolerance for risk, and a higher belief in their own capabilities.

I’ve seen this at work, both in individual leaders and in organizations, and it’s terrifying. Several times as an analyst, I have had to approach my clients to tell them that they have invested in a failing course of action. The treatment for this bias requires three steps:

1.      Recognize escalation behavior as a systemic/causal loop: i.e. loss leads to justification leads to increased commitment leads to increased investment in same cause which leads to loss.

2.      Accept that the continued losses are the result of continued investment in a failing course of action.

3.      Consider alternative courses of action. To reduce the loss, the cause in which the investment is made needs to change.

As a consultant, I am paid to protect my client’s interests, and from time to time that requires an unflattering evaluation of something that my client may feel very passionate about. Sometimes a project, activity, or department represents an escalating loss, and a strong corporate advisor is an effective tool for identifying these biases on a systemic or organizational level.

If you know someone that needs to read this, share the article. If you have any questions, feel free to write to me in the comments, or send me a message.


Markovitch, D. G., et al. (2014). Escalation of commitment in entrepreneurship-minded groups. International Journal of Entrepreneurial Behaviour & Research 20(4).

Pala, Ö., Vriens, D. J., & Vennix, J. A. (2015). Causal loop diagrams as a de-escalation technique. The Journal of the Operational Research Society, 66(4), 593-601.