I recently posted a review of the book Why Can’t You Just Give Me The Number? An Executive’s Guide to Using Probabilistic Thinking to Manage Risk and Make Better Decisions by Patrick Leach. In response, Rick Nason commented about the “Knightian distinction of risk & probability.”
Peter Dizikes wrote an article for MIT News titled Explained: Knightian uncertainty (2 June 2010)
« Frank Knight was an idiosyncratic economist who formalized a distinction between risk and uncertainty in his 1921 book, Risk, Uncertainty, and Profit…. According to Knight, risk applies to situations where we do not know the outcome of a given situation, but can accurately measure the odds.
Uncertainty, on the other hand, applies to situations where we cannot know all the information we need in order to set accurate odds in the first place…“true uncertainty,” as Knight called it, is “not susceptible to measurement.” »
« Some economists have argued that this distinction is overblown. In the real business world, this objection goes, all events are so complex that forecasting is always a matter of grappling with “true uncertainty,” not risk; past data used to forecast risk may not reflect current conditions, anyway. »
« Even so, Knight’s distinction about risk and uncertainty may still help us analyze the recent behavior of, say, financial firms and other investors. Investment banks that in recent years regarded their own apparently precise risk assessments as trustworthy may have thought they were operating in conditions of Knightian risk, where they could judge the odds of future outcomes. Once the banks recognized those assessments were inadequate, however, they understood that they were operating in conditions of Knightian uncertainty — and may have held back from making trades or providing capital, further slowing the economy as a result. »
« Various risks can be classified as Knightian uncertainty. One such risk is reputational risk. We know that all organisations, especially the public ones, face some reputational risk which is a known risk. However, we know little about the impact of reputational risk, which cannot be modelled through conventional risk measurement models. »
« The known unknown risks are probably the most relevant category of risks. It tells us that there is a particular risk that can only be accepted, and there isn’t much we can do about it in quantifying and managing it. »
« Knightian uncertainty, also known as “unknowable unknowns,” refers to risks that cannot be quantified or measured because of a lack of historical data or understanding. This type of uncertainty is different from known risks, which can be modeled and predicted using probability-based tools. »
« It is crucial to understand Knightian uncertainty because it highlights the limitations of risk management models and tools. »
« In economics, Knightian uncertainty is a lack of any quantifiable knowledge about some possible occurrence, as opposed to the presence of quantifiable risk (e.g., that in statistical noise or a parameter’s confidence interval). The concept acknowledges some fundamental degree of ignorance, a limit to knowledge, and an essential unpredictability of future events. »
« Knightian uncertainty is named after University of Chicago economist Frank Knight (1885–1972), who distinguished risk and uncertainty in his 1921 work Risk, Uncertainty, and Profit »
« Work on estimating and mitigating uncertainty was continued by G. L. S. Shackle who later followed up with Potential Surprise Theory. However, the concept is largely informal and there is no single best formal system of probability and belief to represent Knightian uncertainty. Economists and management scientists continue to look at practical methodologies for decision under different types of uncertainty. »
Articles of Interest 
One thought on “Knightian uncertainty”