CFA Conference: Daniel Kahneman

Daniel Kahneman is Professor Emeritas of the Woodrow Wilson School at Princeton and a Nobel laureate in Economics for his work on behavioral economics.  His most recent book is the terrific Thinking Fast and Slow.  His presentation is entitled Psychology for Behavioral Finance.

My session notes follow.  As always, these are at-the-time notes.  I make no guaranty as to their accuracy or completeness.

• Much of our thinking is “fast” — intuitive; System 1.
• System 2 — “slow” thinking takes effort; it covers deliberation and monitoring System 1.
• We are susceptible to priming (upper class British accent talking about large tattoos — we don’t see that as likely and are thus surprised).
• After hearing about losing a wallet in a crowd, we are primed to hear “pickpocket.”
• System 1 interprets the world, not always accurately (tells coherent stories while ignoring ambiguities and alternative explanations).
• We tend to generate the “best” stories and to accept them — not necessarily governed by data.
• Recommends Taleb’s The Black Swan.
• When Saddam captured, stock market rallied (headline links rally to capture); when the market fell later in the day, that was also attributed to the capture.
• System 1 a storyteller and an interpreter of what happens.
• We think of ourselves in System 2 terms — effortful and deliberative.  Much of System 2 is covering up and interpreting System 1 thinking.
• Confirmation bias — we stick with our stories and interpretations as long as we can, even unreasonably.
• The lottery effect and risk aversion both impact us a lot; how these options are framed will greatly impact the results (framing in terms of “mortality rate” causes people not to get treatment while framing in terms of “survival rates” does).
• If we have a good story, we are confident in its truth (and vice versa), regardless of te quality of evidence in support of it — explains a lot of market behavior.
• Coherent stories quash doubts and support inherent overconfidence.
• CFOs of Fortune 500 companies asked about forecasts re S&P 500 in terms of confidence; their forecasts were wildly in error, and they were supposed to be experts (“80% confident” was usually wrong — 67% of the time!); overconfident CFOs took more risks than others in professional lives — overconfidence causes damage.
• System 1 consequence — intuitive rarely believes itself to be stumped (and we’ll then look like crazy to support it, no matter how off-the-wall if is).
• On average, people know very little about the market and don’t predict market movements very well (even though we feel like we know a lot — we’d think we offered value even if/when we didn’t).
• Insidious influence of System 1 inhibits successful analysis.
• Not much we can do about these inherent biases!
• knowing about the errors isn’t a recipe for avoiding them.
• There are few things we can do.
• When we note a risk — slow down.
• Anchoring a big deal, especially in negotiations; so prepare to deal with it.
• Organizations can do more in this area than individuals — quality control (examine systems and processes routinely).
• Meetings — too much weight on those who talk first and talk loudest; helps to ask people to write down their views before the discussion starts — gets better results.
• In organizations — optimism bias a real problem; ask group charged with a decision to write a diary of what went wrong (before the fact) with respect to a preferred approach; that helps decision-making.
• Gossip is important — we talk about the failings of others; Kahneman wants to “educate” gossip — if we anticipate intelligent gossip, we’ll make better decisions.
• Forecasting is poor, but the real problem is that people don’t know the boundaries of their expertise.
• If we know our limits, we could be hopeful.
• Overconfidence is essential;ly inevitable.
• Scenario-thinking is problematic (creates stories and we are vulnerable to good stories).
• We are also prone to superficial judgments too.
• We can know are clients better; we need to make realistic assumptions about what they really want; “risk tolerance” isn’t the way most people think.
• Most people want some mix of “sure thing” and “longshot.”
• Watch out for clients most prone to regret (or keep them in safe assets).
• We should all turn off CNBC and (per Thaler) shouldn’t look at portfolio returns even once a quarter — we react and we react wrongly,  buying high and selling low.
• Discussions with clients are both joint discovery and education; we need clients to imagine their reactions in various scenarios.
• We need reputations as truth-tellers, because some competitor will always tell a better and rosier story.
• We need to condition people for the level of loss they can bear before (wrongly) changing course.
• Neuroeconomics is exciting even though current methods are still primitive.
• We don’t really want to put behavioral controls in place; being systematic helps — records, consistent processes, looking at pieces one-at-a-time and accountability (these things make us very nervous; afraid they will come back to bite us).
• His work was important in economics because of what economists could do with it.