Buy the Book: The book is a collection of Richard Zeckhauser’s maxims for better decision-making. The maxims act as mental models to help understand and simplify a problem, handle uncertainty, and make decisions.
The Notes
- The book is based on a course taught by Richard Zeckhauser called “Analytic Frameworks for Policy” where students learn analytical methods across economics, decision analysis, behavioral decision, game theory, operations research, and more. The key ideas from the course are arranged as “maxims.”
- Tacit knowledge – “that which we know but cannot say.” — Michael Polanyi
- “Complexity could be mastered through the careful application of logic, that strategy should be based on the assumption that rivals would also make strategic choices, and that one often had to make choices knowing that one was more likely than not to be wrong.” — Larry Summers
- “Richard taught me by example that it is easy but not so important to support, congratulate, and be with people when they are up. People need their friends and need new friends when they are down.” — Larry Summers
- “Maxims are useful for two reasons. First, they make us think about things that might not be intuitive at first… Second, they can help us correct behavior when we intuitively know what the right behavior is but yet somehow don’t engage in it.”
- Maxim: When you are having trouble getting your thinking straight, go to an extreme case.
- This maxim is helpful to:
- Understand a concept, problem, or idea better; or
- Assess the best-case or worst-case consequences of a decision you are struggling to make.
- Ex: “Mary and Jim want to paint a room together. If Mary painted alone it would take her 2 hours, and if Jim painted alone it would take him 3 hours. How long would it take to paint the room if they paint together?”
- Most people quickly answer 2.5 hours.
- The extreme case suggests that the answer can not be more than 2 hours. Because if it takes Mary 2 hours to paint the room alone, it should take less than 2 hours if Jim helps Mary paint.
- Marginal Propensity to Consume (MPC) – the proportion of a raise that is spent compared to what is saved. The extreme cases of MPC — 0 and 1 — help explain the concept. An MPC of 0 means the person saves all of it (spends nothing). An MPC of 1 means the entire raise is spent.
- People default to proportional division as a fair outcome in negotiations even when it’s not fair. (Think about dividing a dinner tab equally despite not everyone ordering the same meal).
- Going to an extreme case helps us learn from outliers.
- Using an extreme case can help better understand the risks of a situation. For example, from an investment perspective, a worst-case scenario for a business or portfolio focuses the discussion but can offer a sense of security. If a business or portfolio can survive a crisis, then a non-extreme situation is manageable.
- This maxim is helpful to:
- Maxim: When you are having trouble getting your thinking straight, go to a simple case.
- KISS (Keep It Simple, Stupid) – most systems work best when kept simple instead of made complicated. Unnecessary complexity should be avoided.
- Occam’s Razor – the simplest explanation is usually the correct one.
- Simplifying can help better understand a problem in an overly complex world.
- Maxim: Don’t take refuge in complexity.
- Getting lost in the complexity of a problem avoids thinking about the core of the problem.
- ” While a model can yield a result that might not have been obvious, an analyst’s job is not complete until he or she can decipher the intuition behind the unexpected result – and be able to explain it to decision and policy makers in plain language (maybe with the help of a diagram or two). Models whose results remain a mystery are not useful; models that can be translated into intuitive insights and be broadly understood can be useful.” — Milton Weinstein
- “Appropriate simplification is the great art of modeling.” — Richard Zeckhauser
- ” Too much simplification, and the model will omit key drivers of the phenomenon under study. Too little simplification, and the model will likely be unsolvable. Thus, while the maxim is formulated conservatively in terms of art, it is about much more than art. Appropriate simplification is what modeling is all about, and the maxim provides a useful criterion to judge a model. It is a great guide when reading papers and listening to seminar or conference presentations.” — Alexandre Ziegler
- We naturally default to complicated solutions for complicated problems rather than simpler solutions.
- Why is simple hard to do? Most complex problems are hard to simplify. For most of our life, we’re taught to go against the KISS approach. Schoolwork, for example, never required editing to the simplest points. It was bulk word dumps to meet page quotas.
- Ex: A coach teaching kids how to hit a baseball can give a 50-page slide deck on the science of hitting or he can focus on the core problem — connecting the bat with the ball — and simplify it so kids can understand. The 3 factors are: Hands – hold the bat right, Hips – swivel your hips when you swing, and Head – keep your eye on the ball at all times.
- Rule of Three – a simple principle that suggests things organized in threes are more satisfying, effective, and memorable than in other numbers.
- Maxim: When trying to understand a complex real-world situation, think of an everyday analogue.
- Everyday examples are easier to understand and more memorable and can then be applied to more complex situations.
- Useful in complex negotiations, communicating complex ideas, starting new projects, etc.
- Ex: “Imagine you are contemplating transporting 100 identical items, one by one, over the Niagara Falls using buckets. There are two types of buckets. The first type has been used 100 times and succeeded in 70 of them. The second type has been used 2 times and succeeded only once. What would you do?”
- The second bucket should be studied further, due to lack of data, until there are a statistically significant number of attempts to conclude if its success rate is better or worse than the first bucket.
- The larger the uncertainty around one option, the more likely you need more evidence before making a good decision. But if you discard one option due to uncertainty for something you’re familiar with, you’ll never know if you made the correct choice.
- ” When making decisions, such as whether to invest time and effort in a project, think of its option value. Projects with more uncertain outcomes are likely to be more interesting, and potentially high impact… Higher-uncertainty projects can bring much higher gains.”
- Affordable Concessions – “We should never become so enamored with the strong points in our thinking that we overlook the weakest links in the chain. Furthermore, acknowledging the weakest links is often persuasive when we communicate with others. Concessions convey honesty and objectivity and make what we say more credible.”
- Maxim: The world is much more uncertain than you think.
- “This maxim reminds us that certainty, and near certainty, is an illusion and that the world is a much more uncertain place than most people believe. So the next time you find yourself thinking that some event will happen for sure or that some other event has no chance of happening, pause to remind yourself of this maxim.”
- Most of us view the chance of some event occurring as either certain it will happen or not at all rather than thinking in terms of probabilities.
- Hindsight Bias – we tend to view past events as more predictable than they were. We try to explain away low-probability events as if it were more likely rather than admit that we should expect some low-probability events to occur in any given year.
- “Recognizing that the world is an uncertain place can improve planning.”
- Availability Heuristic – ” Events easily brought to mind are termed ‘available.’ Available events play a significantly disproportionate role in affecting judgments, such as probability assessments.”
- “Just because we do not know what type of high severity event will happen, does not mean that we should not prepare our personal lives, businesses, and investment portfolios to withstand these types of dramatic events. Human psychology leads most of us to extrapolate the near past too far into the future. It lulls us into assurance after periods of calm. However, if we live based on this maxim – that the world is much more uncertain than you think – we will be alert to potential risks and conservatively positioned when these severe events inevitably happen.” — Rich Krumholz
- Governmental policy changes following a natural disaster usually tries to prevent the last disaster — fighting the last war — without considering that the next disaster will not be the same as the last one. Policies need to be broad and robust enough to handle a wide range of surprises. (The same is true of investment portfolios.)
- “Estimating uncertainty is very challenging, even for people who recognize the world is a very uncertain place.”
- Zeckhauser used an estimation exercise that asked students to offer their best guess and the range they were confident the answers to different questions fell into. The exercise shows students how much they underestimate their predictions and beliefs. Students choose too narrow ranges, showing overconfidence.
- “Experience with estimating probabilities reduces but does not eliminate the penchant for underestimating uncertainty.”
- “The core predicament of medicine — the thing that makes being a patient so wrenching, being a doctor so difficult, and being a part of a society that pays the bills they run up so vexing — is uncertainty. With all that we know nowadays about people and diseases and how to diagnose and treat them, it can be hard to see this, hard to grasp how deeply the uncertainty runs. As a doctor, you come to find, however, that the struggle for caring for people is more often with what you do not know than what you do. Medicine’s ground state is uncertainty. And wisdom — for both patients and doctors — is defined by how one copes with it.” — Atul Gawande
- “A policy that depends on accurate oil price forecasts is a policy in trouble…the range of uncertainty is impressive; it is so large that the uncertainty may be the most important feature of the analysis.” — Bill Hogan, energy policy expert, 1985
- Risk, Uncertainty, and Ignorance
- Risk – situations with known probabilities and known states of the world. (Like roulette, we know every possible outcome and the odds to each outcome).
- Uncertainty -situations with unknown probability but known states of the world.
- Ignorance – situations with unknown probability and unknown states of the world.
- “Casinos, which rely on dice, cards and mechanical devices, and insurance companies, blessed with vast stockpiles of data, have good reason to think about risk. But most of us have to worry about risk only if we are foolish enough to dally at those casinos or to buy lottery cards to a significant extent.” — Richard Zeckhauser
- “Uncertainty, not risk, is the difficulty regularly before us.” — Richard Zeckhauser
- “Ignorance is an important phenomenon, I would argue, ranking alongside uncertainty, and way above risk. Ignorance achieves its importance, not only by being widespread, but also by involving outcomes of great consequence.” — Richard Zeckhauser
- Maxim: Think probabilistically about the world.
- Something that has a 22% chance of winning means that if something were repeated 100 times, you should expect to win 22 out of 100 times.
- Subjective Probability – A probability-based partially on judgment because the range of possible outcomes is not entirely known.
- Elements of Thinking Probabilistically
- Understand what subjective probabilities mean.
- Assign probabilities to many events to better understand the world and improve decisions.
- Update those probabilities when we get new info (Bayesian updating).
- To promote probabilistic thinking, replace terms like “most,” “likely,” etc. with numeric estimates.
- Asking open-ended questions vs yes/no questions gets more information that may help improve your estimates.
- “When updating probabilistic beliefs, it is important to use all available information that is relevant, including information from the decisions of others.”
- Ex: “An intriguing example involves purchasing an asset, such as a parcel of land, from a well-informed party. You think the land is undervalued at the proposed price, but the selling party is willing to part with it at that price. You draw the appropriate inference that she does not think it is undervalued.”
- “I now believe that even if people understand probability theory, it is challenging to learn how to apply these concepts when making decisions, particularly when dealing with extreme events, such as natural disasters, few people think probabilistically when deciding whether to protect themselves against future losses.” — Howard Kunreuther
- We don’t fully control the outcome to decisions, so the best question is not “How do I control the outcome?” but “How can I influence the odds?”
- Maxim: Uncertainty is the friend of the status quo.
- Status Quo Bias – People tend to do nothing or maintain previous decisions even when an obvious alternative is better.
- Setting a good default option that works for almost everyone is one way to take advantage of the status quo bias (i.e. auto-enrollment in a retirement plan or organ donation).
- “It is better to teach a person how to decide than to nudge a person to make the right decision.” — Richard Zeckhauser
- Netflix uses status quo bias to keep people binge-watching a show by automatically starting the next episode.
- Action Bias – a tendency to act just be do something even when doing nothing is the best choice.
- Maxixm: Good decisions sometimes have poor outcomes.
- Resulting – Confusing the quality of a decision with the quality of an outcome.
- Decision vs Outcome
- Good Decision and Good Outcome = Decided Well
- Good Decision but Bad Outcome = Unlucky
- Bad Decision but Good Outcome = Lucky
- Bad Decision and Bad Outcome = Decided Poorly
- The quality of decisions should be judged based on what you knew when you made the decision, not what you knew afterward. Judging the process that led to the decision is the first step.
- The best path for uncertain situations is only known after the fact.
- Decision Tree – “When faced with a decision, think about the choices you have and for each of these choices think about what outcomes you associate with each of the scenarios that are uncertain at the moment.”
- “An extremely poor strategy, but frequently taken, is to put off a decision just because it is difficult, even when you do not expect new information to arrive.”
- On average, good decisions should produce more good outcomes than bad outcomes. And a good decision process should improve the chance of good outcomes.
- “Knowing that a good decision can sometimes result in a bad outcome should make us more compassionate with ourselves and with others when bad outcomes occur.”
- “A corollary to this maxim is that bad decisions sometimes have good outcomes.”
- Governmental policy successes should be judged not by how things have changed from point A to point B but by how things are at point B compared to what would have happened without the policy.
- Maxim: Some decisions have a high probability of a bad outcome.
- Like investing in a startup — a high chance of losing it all but a low chance of massive upside.
- Like an experimental medical treatment for a potentially terminal illness.
- Like the choice of nuclear deterrence to avoid nuclear war.
- Maxim: Errors of commission should be weighted the same as errors of omission.
- Errors of Commission – errors made when a person chooses to act but turns out wrong.
- Ex: Buying Amazon stock in 2003, then selling it after it doubles, and missing out on two decades of phenomenal returns.
- Errors of Omission – errors made because a person chose not to or failed to act.
- Ex: Researching Amazon stock, choosing not to buy it in 2003, and missing out on the returns.
- Omission Bias – People weigh errors of commission more heavily than errors of omission. Possibly because we don’t know about the losses from errors of omission (out of sight, out of mind).
- Omission bias helps promote status quo bias.
- Errors of commission generate more short-term regret.
- Errors of omission generate more long-term regret.
- Errors of Commission – errors made when a person chooses to act but turns out wrong.
- Maxim: Don’t be limited by the options you have in front of you.
- “Decisions sometimes require you to reject the options you were given, and to look for others.”
- Sometimes you need to watch out for better options.
- Maxim: Information is only valuable if it can change your decision.
- Good questions to ask: “What might we do differently if we had this information?” If the answer is nothing, the information is not valuable.
- “Don’t wait for information if it won’t change your decision.”
- Maxim: Long division is the most important tool for policy analysis.
- “This maxim makes a simple but critical point: In assessing any policy measure in any realm, you should first determine what it achieves, and what resources it requires. With these tallies, you should then use long division, to compute output per unit of input.”
- Cost Benefit Analysis – a data-driven approach to evaluate potential decisions free from biases.
- “Often the input will be something other than dollars, or dollars are a secondary consideration.” — Richard Zeckhauser
- Output per dollar spent is usually not constant due to diminishing returns.
- “Frequently, long division of output divided by dollar or by some other input (such as one’s time) is just a way to enable clear thinking about an issue. Even if the numbers are hard to calculate, guesstimating the value of the ratio of outputs to inputs tells us if we are in the ballpark of reasonability. Moreover, if the value is high, we might do more of the activity, perhaps in some other locale. If it is low, we should consider cutting the activity.” — Richard Zeckhauser
- “In sum, this maxim involves figuring out some metric of “bang per buck” for different options, ordering these, and then choosing the options with the highest bang per buck until the bang becomes worth less than the resources required, or you simply run out of resources.”
- Simple ratios sometimes offer more meaning and are easier to understand than complex statistical models.
- Maxim: Elasticities are a powerful tool for understanding many important things in life.
- Elasticity – measures how much one quantity may change if another quantity increases by one unit (like 1%).
- Think at the margins.
- Price elasticity of demand – the percentage change in the amount demanded of a good when the price increases by 1%, all else equal.
- Betrayal Aversion – people view losses due to betrayal as worse than losses due to bad luck. Ex: giving a $100 to a friend who skips town and never pays you back is viewed as worse than losing $100 because of a hole in your pocket.
- Elasticity of Trust – measures how much people are willing to trust given changes in the cost or likelihood of betrayal.
- Maxim: Heterogeneity in the population explains many phenomena.
- “Average behavior” is rarely the norm in large groups.
- The key to analyzing large-group behavior is to break the group into smaller homogeneous segments with similar behaviors and analyze each segment.
- “Understanding heterogeneity frequently helps to diagnose problems where behavior changes over time.”
- Maxim: Capitalize on complementarities.
- Complementarities are positive cross-partial derivatives — where an increased quantity in one input makes an increased quantity of another input more valuable.
- Ex: If you own a tennis school business, with one tennis court, there’s a point where adding one more tennis instructor becomes less valuable to your business. But if you add one more tennis court, the value of having another instructor rises.
- Ex: Collaboration – collaborating with someone with a similar skillset as you is often less valuable than if you collaborated with someone who brings a different skillset to the table. A theorist gains more from collaborating with an empiricist than another theorist.
- “Positive cross-partials are the hallmark of human existence. On our own, we are puny creatures who would fare poorly against any large carnivore in the wild. Our individual actions are only effective because they are supported by the actions of thousands of others – today and through history.” — Ed Glaeser
- Maxim: Strive not to be envious — see your friend’s success as your gain.
- “Envy is one of the most potent causes of unhappiness.” — Bertrand Russell
- Maxim: Eliminate regret.
- Avoid regret when a good decision turns out poorly. A good decision is no less good because of an unlucky outcome. Learn what you can from it and move on.
- Corollary: Avoid pride in poor decisions that turn out well.
- “In sum, regret is a wasteful emotion regardless of whether the bad outcome is a result of a good or a bad decision. Realizing you made a mistake and taking action to correct it is good, but dwelling on the mistake does not do any good. Perhaps even worse, our desire to avoid regret in the future might lead us to make suboptimal decisions in the present.”
- Regret often leads to overweighting errors of commission over errors of omission leading to status quo bias. Feeling regret is stronger when bad outcomes are due to a choice rather than doing nothing. It can lead to do nothing next time — to avoid regret again — when taking action is the best choice.
- Regret avoidance leads to the sunk cost fallacy.
- Sunk Cost Fallacy – Occurs when we use prior investment of time, money, or resources to justify investing more time, money, or resources into something. Throwing good money after bad. It avoids potential regret felt from a decision that turned out poorly.
- Maxim: Make pleasure-enhancing decisions long in advance, to increase the utility of anticipation.
- “Anticipation of a future reward, such as an upcoming vacation, can be sometimes even more gratifying than the experience itself. This is the idea behind this maxim and what some people have termed anticipation utility.”
- Try to maximize the pleasure and utility of decisions already made. Make the best of it. Think about the positives.
- Corollary: Avoid thinking long and often about upcoming negative events.
- Life Maxims:
- There are some things you just don’t want to know.
- If you focus on people’s shortcomings, you’ll always be disappointed.
- Practice asynchronous reciprocity.